|UFDC Home||myUFDC Home | Help|
This item has the following downloads:
DEVELOPMENT OF A COMPETITIVE INHIBITION ENZYME-LINKED
IMMUNOSORBENT ASSAY (CI ELISA) FOR SEROSURVEY OF WILDLIFE
SPECIES FOR WEST NILE VIRUS EMPHASIZING MARINE MAMMALS
A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE
UNIVERSITY OF FLORIDA
This work is dedicated to my mother, for instilling me with the tenacity I needed, to my
father for infusing me with a love of learning, and to my wonderful loving husband who
has always been there to support me in every way possible. And lastly, a special
dedication to my daughter Emily, who makes every day a joy.
I would like to thank all of the institutions, biologists, veterinary technicians, and
veterinarians who helped me in collecting all the serum samples for this project. Personal
thanks go to Bob Bonde and Mark Sweat for giving me access to many archive serum
samples. Thanks also go to the staff at the marine mammal pathobiology lab, USGS, Sea
World Orlando, Miami Seaquarium, Six Flags Ohio, Columbus zoo, and Lowry park zoo.
In addition I would like to thank Dr. Elliot Jacobsen for alligator samples, Dr. Ramiro
Isaza for elephant samples, Dr. Darryl Heard for bat samples, and Dr. Cynda Crawford
for cat samples. Big thanks go to all the researchers and volunteers involved in the
manatee captures. Without their help I would have been unable to obtain most of my
I would also like to thank the members of Dr. Long's lab. They helped me with my
research and made it a welcome and enjoyable place to work. Thanks go to Sally
O'Connell for helping me figure out everything that I was supposed to be doing. I would
also like to thank the staff at the marine mammal path lab. They graciously allowed me
the time to finalize my thesis. I give special thanks and acknowledgment to Dr. Butch
Rommel for allowing me to use his wonderful anatomical diagrams of nerves, and for his
help with revisions. Lastly, but certainly not least, I would most graciously like to thank
my advisor, Dr. Maureen Long for her help. She took me on without completely
realizing what she was getting into, but has remained a dedicated and inspirational
advisor. It was ajoy, an honor, and a wonderful learning experience to be her first
TABLE OF CONTENTS
A C K N O W L E D G M E N T S ................................................................................................. iv
LIST OF TABLES ...... .... ................................. .......... ............ .. ix
LIST OF FIGURES ................................. .. ... ... ................. .x
ABSTRACT ........ ........................... .. ...... .......... .......... xii
1 IN TRODU CTION ................................................. ...... .................
2 LITER A TU R E REV IEW .............................................................. ....................... 4
General Causes of Neurological Disease in Marine Mammals.............. ...............4
B acterial and V iral Encephalitides .................................. ............ .................. 5
B ru c e lla ...................................................... ................ 5
M o rb illiv iru s..................................................................................... 6
R ab ies ............................................................... ... .... .. ..... 7
A rb ov iru ses ....................................................... 8
Parasitic Encephalomyelitis................... .......... .......................... 9
Toxoplasmosis........................ ................... ...............
S arco cy stis ............................................ 9
N a sitre m a ............................................................................................... 1 0
Neurotoxicosis ...................................... 10
Paralytic shellfish poisoning ........................................... 10
Brevetoxin ................................................................ ..... ............ 10
D o m o ic a c id .............................................................................. 1 1
N eurological Evaluation ........................................................ ...................... 11
D diagnostics ................................................................. ............................ 11
N eu rolog ical E x am ........................................................................................ 12
Cranial Nerves .................................... ......... ................... 13
W est N ile Virus .......................................... 19
T ax o n o m y .....................................................................................1 9
L ife C y cle ................................................................2 0
M o rp h o lo g y ................................................................................................... 2 3
History .......................... ..... ....................25
Pathogenesis and C clinical Signs ................................................................... 25
H u m a n s ............... ...... .......... ................ ........................ 2 6
Horses ..... ........... .... ......... ................27
B ird s ..........................................................................2 8
O th er sp ecies ............................................................2 8
E p id em biology ...............................................................2 9
M o sq u ito e s ............................................................................................. 2 9
Humans .......................................................................... 30................ .........30
B ir d s .............................................................................................. 3 1
H horses .........................................33............................
O th er sp ecies ............................................................34
C lin ical D iag n o sis ......................................................................................... 3 5
Treatment and Prevention.............................. ...............37
3 DEVELOPMENT OF A COMPETITIVE INHIBITION ENZYME-LINKED
IMMUNOSORBENT ASSAY ..................................................... 39
In tro d u ctio n .......................................................................................3 9
M materials an d M eth od s ......................................................................................... 4 0
R agents .................................. ...... ......... ........................ .......... 41
Plaque Reduction N eutralization Test ................. ............................................. 42
Competitive Inhibition Enzyme-Linked Immunosorbent Assay .........................42
Optimization Procedures ................. ...................................44
M onoclonal optim ization ....................................................... 44
Antigen and monoclonal optimization ................................... ....44
Statistical A naly sis ...........................................................4 5
R esu lts ..............................................................................................4 5
Monoclonal Optimization................ .............. .........47
Antigen and Monoclonal Antibody Optimization .............................................48
Statistical A naly sis ...........................................................49
D isc u ssio n ............................................................................................................. 5 7
4 USE OF A COMPETITIVE INHIBITION ENZYME-LINKED
IMMUNOSORBENT ASSAY FOR DETECTING WEST NILE VIRUS
EXPOSURE IN A POPULATION OF CAPTIVE BATS .......................................63
In tro d u ctio n .......................................................................................6 3
M materials an d M eth od s ......................................................................................... 64
T e stin g ................................................................6 4
R e su lts ...........................................................................................6 5
C o n c lu sio n s........................................................................................................... 6 5
5 S U M M A R Y ................................................................................................6 7
A TESTING OF VERO CELL ANTIGEN ON THE CI ELISA ..............................71
B OPTIMIZATION OF THE CI ELISA USING THE CHIMERIVAX ANTIGEN.73
A n tig en D ilu tio n s............................... ........................................................... .. 7 3
M onoclonal A antibodies .................................................... .............................. 74
Com prison of Conjugate D iluents ........................................ .............................. 74
P ro c e d u re ....................................................................................................... 7 5
R e su lts ........................................................................................................... 7 5
N negative Control ............... .... ............................................ ...... .... 76
Second Chim eriVax Antigen Batch ........................................ ..... ............... 77
L IST O F R E F E R E N C E S ....................................................................... ... ................... 79
BIOGRAPHICAL SKETCH ............................................................. ............... 102
LIST OF TABLES
3-1 Plaque reduction neutralization results by species .................................................46
3-2 Agreement between cat samples on the CI ELISA and the PRNT ........................53
3-3 Disagreement between PRNT results and CI ELISA results of the elephant
sam ples ..................................... ................................... ........... 56
4-1 Sample breakdown of the bat species tested ............................. ..................64
4-2 Competitive inhibition ELISA results on bats tested showing a large percentage
of positive sam ples ..................................... ........ .. ...... ........ .... 65
LIST OF FIGURES
2-1 Size comparison of cranial nerves among different mammal species and with the
dog ........... ....... .. ....................................... ................ ................ 14
3-1 Comparison using MAbs 3A3, 5H10, and 7H2 separately and as a group..............47
3-2 Determining optimal 7H2 dilution by testing several dilutions on the CI ELISA...47
3-3 Mouse-brain antigen at two separate dilutions tested against several dilutions of
the 7H 2 m onoclonal for reactivity ........................................ ....................... 48
3-4 ChimeriVax antigen serially diluted and reactivity tested against positive,
negative, and 7H 2 controls............................................... ............................ 49
3-5 An ROC curve demonstrating the sensitivity and specificity of the CI-ELISA
using an OD cut-off value .............. ........................................50
3-6 Separation of the positive WNV animals and the negative WNV animals..............50
3-7 An ROC curve demonstrating the specificity and sensitivity using a percent
inhibition cut-off value .............................................. .... .... .. ............ 51
3-8 Separation of the positive animals and the negative animals using the 28%
inhibition cut-off value .............................................. .... .... .. ............ 51
3-9 Distribution of PRNT-tested positive and negative horses tested on the CI
ELISA and compared to a 28% inhibition cut-off value........................................52
3-10 Results of known positive and negative alligator plasma samples as tested on the
C I E L IS A .......................................................................... 5 3
3-11 Non-manatee marine mammal samples as tested on the CI ELISA using the
percent inhibition cut-off.................................... ........................... .............. 55
3-12 Level above a 28% cut-off for each manatee tested........................................ 55
A-i Results obtained using the Vero cell antigen compared to the mouse-brain
an tig en ..............................................................................7 2
B-l Competitive inhibition ELISA plate coated with the ChimeriVax antigen and
tested fo r reactive ity .................................................................... .... .. ... .. .... 7 3
B-2. Competitive inhibition ELISA plate coated with the ChimeriVax antigen and
tested for reactivity against monoclonals 7H2, 3A3, and 5H10.............................74
B-3 Testing the NVSL positive control at varying dilutions to compare diluting the
conjugated antibody in a blocking solution versus using PBST alone...................76
B-4 Testing the NVSL negative control at varying dilutions to compare diluting the
conjugated antibody in a blocking solution versus using PBST alone...................76
B-5 Comparison of known positive serum samples and their percent inhibition using
the new negative control and the new recombinant antigen .................................77
Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science
DEVELOPMENT OF A COMPETITIVE INHIBITION ENZYME-LINKED
IMMUNOSORBENT ASSAY (CI ELISA) FOR SEROSURVEY OF WILDLIFE
SPECIES FOR WEST NILE VIRUS EMPHASIZING MARINE MAMMALS
Chair: Maureen T. Long
Major Department: Veterinary Medicine
West Nile virus (WNV), an arthropod-borne virus introduced to the New World in
1999, causes neurological disease primarily in birds, horses, and humans. This virus is
maintained in a life cycle involving birds and mosquitoes. West Nile virus has clinically
affected many wildlife and domestic species including harbor seals, monk seals,
alligators, gray squirrels, alpacas, llamas, a wolf, and one dog. Many more mammals
have seroconverted to WNV without any signs of clinical disease. Studies of actual
disease status in wildlife species are lacking.
As of 2005, WNV has been detected in all of the continental United States and has
expanded into Canada, Mexico, and the Caribbean. Currently, several serological
methods are used to identify WNV virus exposure in animals. These assays are either
labor intensive or require the use of species-specific reagents. Our goal was to develop
and validate a WNV-specific ELISA using monoclonal antibodies in a competitive
inhibition format (CI ELISA). Samples were screened using the plaque reduction
neutralization test (PRNT) and compared against the CI ELISA. A total of 262 serum
samples were tested representing 28 different species. Seven species were positive for
WNV including alpacas, camels, cougars, elephants, lions, tigers, and a box turtle. Four
species were positive against SLEV including camels, elephants, a lion, and a tiger.
WNV ChimeriVax antigen was selected over mouse-brain antigen for the CI
ELISA. The use of a single monoclonal (7H2) was determined to be as efficient as using
a combination of three. Inter-assay variation based on a calculated coefficient of variance
(CV) was 7.9% with a standard deviation of 4.9%.
Alligators and horses samples were analyzed to determine a proper cut-off. Using a
receiver operating characteristic (ROC) curve analysis, a cut-off of 28% inhibition was
identified, resulting in an assay sensitivity of 92.6% and a specificity of 100.0%. The
alligator data resulted in 100% specificity and sensitivity. The best cut-off for elephants
was 0% inhibition, which essentially renders the CI ELISA inappropriate for this species.
Sample numbers were inadequate to validate this assay for any marine mammal, cat, or
Results showed that this assay has higher specificity than sensitivity. Using the
ROC curve, this value can be increased by changing the cut-off value; however, the
specificity of the test would decrease. As the PRNT is the gold standard, confirmatory
testing would be used to validate the CI ELISA results in a diagnostic laboratory setting.
Results indicate that this ELISA can accurate identify WNV exposure in alligators and
horses. Our assay would need to be validated for each individual species and cannot be
used as a broad-based screening assay.
West Nile virus (WNV), an arthropod-borne virus introduced to the New World in
1999, causes neurological disease primarily in birds, horses, and humans.1 Since its
initial appearance, this virus has spread rapidly across the North American continent.2 In
2002 alone, over 15,000 equine cases were reported by the U.S. Department of
Agriculture (USDA).2 Thus far, harbor seals, monk seals, alligators, gray squirrels, an
alpaca, a llama, a wolf, and one dog have all been clinically affected by natural
infections.3-10 A wide range of other animals have seroconverted, with unknown clinical
status. In the late summer and fall of 2001, WNV became endemic in Florida. During
2002, WNV was established in much of the U.S. affecting over 2000 people with
WNV encephalitis, and 2000 with WNV fever in 43 states. In addition, 14,717 horses
and 137 species of birds were affected. The actual infection/disease status has not been
completely assessed for many species. Because of the presence of this virus in many
species of birds and mammals, assessment of multiple species of free-ranging and captive
wildlife species is needed to determine the danger of WNV infection for Florida's
valuable wildlife resources.
Florida marine mammals may be at particular risk for arboviruses, since these
viruses have been described in several aquatic species. Alphaviruses have recently
appeared in aquatic fish species such as Atlantic salmon and rainbow trout, indicating the
possibility of aquatic transmission.1112 Pollution and habitat destruction are thought to
make certain species of marine mammals susceptible to acquisition of new viruses and
increased outbreaks of marine mammal-specific viruses.13 Recently, an alphavirus,
similar in morphology to Eastern equine encephalomyelitis (EEE) virus was detected in
southern elephant seals (Mirounga leonina) in Australia.14 Transmission is thought to
occur through blood sucking lice. In 2002, a captive harbor seal (Phoca vitulina) in New
Jersey was diagnosed with WNV, making it the first marine mammal reported to die from
this disease.5 Since then, there have been several other reports of P. vitulina and
Hawaiian monk seals (Monachus schauinslandi) clinically affected by WNV.3'6
St. Louis encephalitis virus (SLEV) is a flavivirus like WNV, and is endemic in the
state of Florida.15 Any serosurvey that involves WNV must also include SLEV, since
these have the potential to cross-react on testing.16 Because of the multiplicity of
syndromes, infection of arboviruses could easily be overlooked.
The gold standards for serologic testing for encephalitis viruses are neutralization
tests. These tests are applicable to multiple species, but are expensive as screening tools,
requiring viral expertise, special facilities, and often special permits. The availability of
WNV monoclonal antibodies allowed development of an ELISA in a competitive
inhibition (CI ELISA) format to test multiple species. This CI ELISA was developed as a
rapid screening assay for use in multiple wildlife species.
The purpose of this study was to examine WNV and SLEV seroprevalence in
Florida mammals (with an emphasis on marine mammal species), and to develop a CI
ELISA for broad-based species testing. Several specific aims were identified.
* Specific aim 1: Perform a serosurvey of wildlife species in the state of Florida using
the plaque reduction neutralization test (PRNT).
* Specific aim 2: Develop a CI ELISA using known positive and known negative
* Specific aim 3: Use the CI ELISA to perform screening of blinded samples.
* Hypothesis: Wildlife species in the state of Florida are being exposed to WNV.
Using the CI ELISA will allow broad-based cross-species screening for WNV.
General Causes of Neurological Disease in Marine Mammals
Three orders were emphasized in our review of captive and free-ranging marine
* Cetacea: whales, dolphins, and porpoises
* Carnivora: with suborder Pinnipedia including seals, sea lions, and walruses; and
the families Mustelidae and Ursidae consisting of the sea otter and the polar bear
* Sirenia: manatees and dugongs
Encephalitic diseases in marine mammals are caused by numerous etiologic
agents. These include parasites, bacteria, viruses, neoplasias, and toxins. As clinical and
diagnostic marine mammal medicine and research progresses, new etiologic agents are
being discovered. Many of these agents have been around for some time; others appear
to be emerging. Historically, neurological diseases in marine mammals are not
uncommon. One study on stranded cetaceans off the coast of Italy from 1990 to
1997 found that encephalitis was present in nearly 11 of 17 (64%) of the striped dolphins
(Stenella coeruleoalba) in which the brain was examined.17 However, in another study in
Belgium and France, only 6 of 55 stranded harbor porpoises (Phocoenaphocoena)
examined had evidence of an encephalitis.18 This may be partly due to the fact that
certain species are more susceptible to certain diseases than others. Thus far, a variety of
viruses, bacteria, and protozoa have been identified as causative agents for neurological
diseases in marine mammals. These are discussed next.
Bacterial and Viral Encephalitides
Brucella is a Gram-negative, facultative, intracellular bacteria that affects a wide
range of species, including humans. Infection in marine mammals was first reported in
1994 in free-ranging seals and cetaceans.19'20 Recent PCR studies of the Brucella species
affecting marine mammals indicate that there are two (newly discovered) species; one
type affecting cetaceans and the other affecting pinnipeds.21 The proposed new naming
scheme is B. cetaceae and B. pinnipediae respectively. To date, there have been no
reports of Brucella in Sirenia.
A large number of serosurveys have been performed to detect the presence of
Brucella antibodies in free-ranging marine mammals. A large number of members of all
families of cetacean and pinniped species have tested serologically positive for Brucella
exposure.19,22-28 Polar bears (Ursus maritimus) have also tested serologically positive for
Brucella although there has been no report of clinical disease.29
Although serological evidence of exposure has been found in many species,
clinical disease has been rarely described. Two captive bottlenose dolphins (Tursiops
truncatus) experienced a placentitis and abortions and the fetuses had evidence of a
Brucella infection.30 Neurological disease associated with Brucella in marine mammals
was described when a chronic, non-suppurative meningoencephalitis was found in three
young S. coeruleoalba.31 As these animals were found dead, no description of the
clinical syndrome has yet been reported.
The exact means by which Brucella is transmitted among marine mammals
remains unclear. There are several hypotheses including casual contact, sexual contact,
maternal transmission, trauma, ingestion, and parasites.27 There is a zoonotic concern
since many live marine mammals can become beached and infected carcasses could wash
up on shore. Two cases of human neurobrucellosis have been reported with
B. pinnipediae. The method of transmission remains unclear, as both subjects denied
contact with any marine mammals.32
Thus far, the need has not arisen to treat captive animals. The marine mammal
B. pinnipediae form has been treated successfully in a human using a six-week course of
rifampin and doxycycline.33
Morbillivirus is a single-stranded RNA virus in the family Paramyxoviridae.
Until recently, there were four genus members: measles virus (MV), canine distemper
virus (CDV), rinderpest virus (RPV) and peste-de-petits ruminants virus (PPRV).34'35 In
1988, a series of epizootics occurred that led to the discovery of several previously
undescribed morbilliviruses specific to marine mammals.3638 These include phocine
distemper virus (PDV), dolphin morbillivirus (DMV), and porpoise morbillivirus (PMV).
Marine mammals are currently affected by four members of the Paramyxoviridae family:
CDV in seals39 and polar bears,40'41 PDV in seals,42 DMV in dolphins and whales,43'44 and
PMV in porpoises.45 DMV and PMV are closely related antigenically whereas PDV is
closely related to CDV.43
There have been at least 8 recognized epidemics attributed to morbillivirus since
1987 encompassing North America, Europe, Asia and Africa and including numerous
cetacean and pinniped species.34'35'38'42'46'46-54 Although bronchial pneumonia and
alveolitis are the most common findings, there can be CNS disease present with
morbillivirus infections. The cerebrum is most commonly affected and
histopathologically, a neuronal necrosis, gliosis, perivascular cuffing, and demyelination
with astrocytosis and syncytia occurs.34'35 Clinical signs reported in pinnipeds include
depression, muscle twitching, abnormal posture, head tremors, convulsions, seizures, and
increased tolerance to humans.34'35'55 Clinical signs in cetaceans have been observed less
frequently but have included respiratory difficulty and abnormal swimming.34'55
Diagnosis can be based on characteristic histopathological lesions.35 Other methods such
as neutralization tests, paired rising titers, and ELISA tests are also used.34'35'56'57
Treatment for morbillivirus-infected marine mammals is primarily supportive and
the mortality rate is high. Vaccination has been performed in Europe using various
commercially available canine distemper vaccines including a modified-live canine
distemper, a killed, and a subunit vaccine. These vaccines appear to elicit a protective
antibody response, but there have been no studies to verify efficacy in preventing disease
or degree of virus shedding in the vaccinated animals. Most North American zoos and
aquaria have strict quarantine procedures to prevent morbillivirus from entering their
facilities. As such, vaccination is not performed in North America.58
Rabies virus, a fatal cause of encephalitis, is a member of the family
Rhabdoviridae. Inoculation by bite is the primary mode of transmission and many
mammalian species are susceptible to development of disease. Although all marine
mammals are assumed susceptible to the rabies virus, there have been relatively few
reports of marine mammals contracting the disease. In 1981, a ringed seal (Phoca
hispida) in Norway was found disoriented and wounded.59 The seal's condition
progressively worsened and it became aggressive. Rabies was later confirmed using
immunofluorescence. The seal likely may have been infected by contact with rabid
Hunters reported a polar bear with unilateral hind limb paresis.60'61 Following its
death, histopathology of the lumbar spinal cord revealed moderate to severe mononuclear
inflammatory cell cuffing and gliosis in the gray matter. Rabies was later confirmed via
immunoperoxidase testing of the spinal cord and a positive mouse inoculation test. Negri
bodies were not detected in the spinal cord and there were no lesions in the brain. The
bear was likely infected through contact with an infected canid in the area.
Arboviruses stand for arthropod borne viruses and are maintained by a cycle
between vertebrate hosts and hematophagous arthropods such as ticks or mosquitoes.
Arbovirus is not a taxonomic term, but rather, refers to four virus families that rely on
arthropod transmission as part of their life cycle.62 The four virus families include
Togaviridae (including the alphaviruses), Reoviridae, Bunyaviridae, and Flaviviridae.
Historically, no arboviruses had been described in any aquatic species. Recently
however, there have been reports of alphaviruses in rainbow trout (Oncorhynchus
mykiss),63 Atlantic salmon (Salmo salar),64 and southern elephant seals (Mirounga
leonina).14 This latter finding is of particular interest since the virus was isolated from
the elephant seal louse (Lepidophthirus macrorhini) which routinely infests M. leonina.14
It has been suggested that this new virus is transmitted by the louse, as the island on
which infected seals were found has no mosquitoes and the seals are only rarely affected
with ticks.14 There is also a report of a captive killer whale (Orcinus orca) clinically
affected with SLEV which later died of the disease.65 Confirmation was made using
virus isolation. More recently, several harbor seals (Phoca vitulina) and a Hawaiian
monk seal (Monachus schauinslandi) were fatally affected with WNV.3'5'6
Toxoplasma gondii is a commonly reported infection in marine mammals.
Various syndromes have been reported in T. truncates,66'67 sea otters (Enhydra lutris
nereis),68,69 California sea lions (Zalophus californianus),70 beluga whales
(Delphinapterus leucas),71 harbor seals (P. vitulina richardsi),72 and a Florida manatee
(Trichechus manatus latirostris).73 Transplacental transmission is well-documented in
terrestrial animals and T. gondii has been isolated from the tissues of a stillborn late-term
fetal Indo-Pacific bottlenose dolphin (Tursiops aduncus).74 A dual Sarcocystis neurona
and T. gondii infection was reported in E. lutris.75 This sea otter had a severe
meningoencephalitis and malacia with hemorrhages, inflammation, and perivascular
cuffing. Tissue cysts were visible at the periphery of the lesions, with central areas of
necrosis. Diagnosis for protozoal diseases is based on rising titers. Treatment is
primarily supportive although clindamycin has been used successfully to treat M
S. neurona is the causative agent of equine protozoal myeloencephalitis and is
transmitted in the feces of opossums (Didelphis virginiana).77 Despite its terrestrial
nature, there have been several reports of marine mammals being clinically affected by S.
neurona including harbor seals (P. vitulina richardsi)72'78 and northern and southern sea
otters (E. lutris kenyoni, E. lutris nereis).75 79-s In sea otters, isolation of the agent has
been associated with lesions in the CNS resulting in a disseminated nonsuppurative
meningoencephalomyelitis.79 Studies have demonstrated that otters sampled near areas
of maximal freshwater run-off in California have a higher seroprevalence to
Toxoplasma.82 This has led to a hypothesis that marine mammals are becoming exposed
to these protozoa due to storm run-off.68'82
Nasitrema is a trematode that is often found in the nasal sinuses of cetaceans.
Trematode brain lesions have been implicated as the cause of encephalitis in
S. coeruleoalba,83 common dolphins (Delphinus delphis),84 Risso's dolphins (Grampus
griseus), and Pacific whitesided dolphins (Lagenorhynchus obliquidens).83-5 It has been
speculated that aberrant migration of the trematodes through nerves may result in loss of
auditory function of in dolphins and resultant strandings due to the inability to catch food.
Paralytic shellfish poisoning
Paralytic shellfish poisoning (PSP) is most commonly associated with saxitoxin
but can be caused by other related biotoxins. The toxin accumulates within shellfish that
have consumed dinoflagellates. Consumption of the shellfish results in PSP. Deaths
from PSP have been documented in E. lutris nereis,86'87 humpback whales (Megaptera
novaeangliae),s8 and Mediterranean monk seals (Monachus monachus).89 Clinical signs
reported were lethargy, paralysis, and incoordination.90
Like shellfish poisoning, brevetoxins also accumulate in shellfish and seagrasses
and can cause poisonings in mammals that eat contaminated food items. Ingestion can
result in gastrointestinal and neurological signs although inhalation of the toxin has been
implicated as another route of exposure.91 Manatees (T. manatus latirostris) in
Southwest Florida have frequently been affected by brevetoxin, with severe congestion of
the nasopharynx, lungs, liver, and kidney noted.91'92 The neurological signs reported in
manatees included incoordination, muscle fasciculations, and a loss of the righting
reflex.91 Death often follows. Supportive treatment has resulted in recovery once
animals were removed from the brevetoxin contaminated environment.
Marine diatoms such as Pseudonitzschia are responsible for the production of
domoic acid.93 Domoic acid, when ingested, can have deleterious effects on marine
mammals as well as humans. Experimentally, it has been associated with the
development of neuronal degeneration in rats,94 and has been implicated as the cause of a
severe mortality event in California sea lions in 1998.95 The affected sea lions exhibited
neurological signs which included ataxia, depression, and seizures. The brain lesions
observed in the sea lions were characterized by zonal vacuolization of the hippocampal
neurophile, and were most severe in the ventral hippocampus.95'96 Affected animals had
high hematocrit levels, an eosinophilia and high creatine kinase levels.97
The basic work-up for any neurological patient includes a complete blood panel
including hematology and a serum chemistry profile. This identifies any electrolyte
imbalances, identifies liver disease if present, and indicates inflammatory processes.
Radiography or computed tomography (CT) is possible on marine mammals, depending
on their size, to identify spinal or brain lesions. There have been no reports of
intervertebral disc disease in any marine mammal species although diskospondylitis has
been reported in T truncatus.98 The dolphin did not demonstrate any neurological signs.
Cerebrospinal fluid (CSF), for cytological examination, can be obtained from pinnipeds
(both phocids and otariids) from the atlanto-occipital joint in a method similar to that of
dogs.76 Although unpublished, CSF can also be collected from otters and polar bears in a
similar manner. No method has been published concerning CSF collection in cetaceans
There has been very little published concerning the clinical neurological system
of marine mammals. Work has been done examining the brains of several species,
however, a thorough description of a proper neurological exam in these animals has not
The "classic" neurological exam, as performed in small animals, is rarely possible
when dealing with large non-domestic species because very few animals can be safely
handled without sedation. Although some reflexes remain in sedated animals, a proper
workup is simply not possible. When dealing with marine mammals, this situation is
made more difficult because these animals spend a good portion of their time in a
weightless environment that is, for the most part, inaccessible to us. However, a good
understanding of the nervous system of marine mammals can aid a clinician in assessing
the neurological status of animals in their environment. This section of the thesis will
discuss the similarities and differences of the neurological system of marine mammals,
and how this information can be used to assess a patient. The marine mammals discussed
will include polar bears, otters, manatees, cetacea (both mysticete and odontocete), and
pinnipedia (both phocids and otariids).
Neurological disease is suspected when abnormalities of mentation, posture, and
gait are noted. In marine mammals this would include incoordination, lethargy, paresis,
paralysis, depression, muscle twitching, abnormal posture, abnormal swimming or ataxia,
loss of the righting reflex, etc. For cetaceans and sirenians, all neurological observations
must take place with the animals swimming in the water, therefore access to an
underwater viewing area is helpful. Depressed or lethargic animals may be easier to
examine as they will linger near the surface. Certain non-neurological diseases, such as a
one-sided pneumothorax, can result in an animal listing to one side. Although this can be
a neurological sign, it should not be automatically assumed without further investigation.
Depressed or stuporous animals often indicate cerebral cortical disease or brainstem
Pinnipeds, otters, and polar bears spend part of their time on land where posture
and gait can be easily assessed. This would include such signs as a head tilt, ataxia,
and/or circling. In small animals, postural reactions such as wheelbarrowing and
conscious proprioception are normally tested; however, this cannot be accomplished with
marine mammals due to either the absence of limbs or the presence of teeth. Therefore, it
is imperative that the animal be observed moving in their environment over a period of
time. Reliance on well trained animal keepers can aid in the observation of captive
animals. Loss of muscle tone, if severe enough, can be evaluated visually in non-sedated
animals. When examining an animal under sedation, muscle tone should be evaluated.
Although the polar bears, otters, and pinnipeds likely possess the same spinal reflexes as
small animals, handling prohibits their complete evaluation. In those animals with
paresis and/or paralysis, deep pain should be assessed if at all possible to determine the
Cranial nerve function should be assessed in marine mammals. Although there
are many similarities between terrestrial and marine mammal species, there are several
significant differences which should be noted. Many cranial nerves can be partially
assessed by observation alone. On animals that can be handled, a more accurate
assessment can be made.
A Mustehdae (Otter)
B Sirenia (Manatee)
C -c ( t
C Phocids (Earless seals)
F Otarnds (Eared seals)
D Ursidae (Polar bear)
G Odontocetes (Toothed whales)
H Mysticetes (Baleen whales)
Figure 2-1. Size comparison of cranial nerves I, II, V, VII and VIII among the different
mammal species and with the dog. Used and adapted with permission from
Reynolds JE, Rommel SA, eds. "Biology of marine mammals". 1999;
Washington, DC: Smithsonian Institute. Figure 2-25 on page 60. S.A.
Cranial nerve 0 is the terminal nerve. It was only discovered after the other
nerves were named. It was labeled zero to follow the anatomical naming scheme. This
nerve is present in several species of cetaceans.104'105 These terminal nerves have been
found to contain gonadotropin-releasing hormone (GnRH).106,107 This has led to the
hypothesis that this nerve may be involved in chemoreception for sexual reproduction
however the function of this nerve is still unknown. There is currently no recognized
method to test it properly.
Cranial nerve I is the olfactory nerve. It is interesting to note that this cranial
nerve is completely absent in the odontocete species although they still retain an olfactory
lobe.108'109 As indicated by Figure 2-1, cranial nerve I size varies among different species
(with the polar bear much larger than phocids, for example). This is likely due to the fact
that polar bears rely heavily on smell to hunt.110 Traditionally, the olfactory nerve is
tested in small animals by the use of a noxious odor such as alcohol. In marine mammals
this may be difficult if not impossible. If a lesion is suspected in a captive animal, food
can be hidden in the enclosure prior to allowing the animal allowed to enter. Testing of
this nerve is very subjective. The reduced size of this nerve in many of the marine
mammal species suggests that it is not as important a sense for them as it is for terrestrial
Cranial nerve II (CN II), the optic nerve, is responsible for sight. The optic nerves
of odontocetes and sirenians are much reduced in size compared to other marine
mammals.110 Their reliance on sight is less than the other species and therefore, their
optic nerve function is more difficult to assess. Odontocetes possess echolocation and
captive animals may be able to move around their tank with little difficulty even with a
damaged optic nerve. Sirenians possess sensory tactile hairs on their body which have
been hypothesized to function in a manner similar to the lateral line system of teleosts.111
Pinnipeds possess well developed vibrissae that have been shown experimentally to allow
them to identify objects, even when their vision was completely inhibited experimentally
with eye cups.112 Therefore, observation alone should not be used to evaluate
CN II function in these species.
In animals that can be handled, signs indicating loss of function include a dilated
pupil and the loss of the pupillary light reflex. In most other mammals, the eyes project
the majority of their nerve fibers to the contralateral cerebral hemisphere but there are
still a good number of fibers that project to the ipsilateral side. This results in the
observation of pupillary dilation of the contralateral eyes during examinations. This does
not appear to be the case in dolphins examined both anatomically and
physiologically.113-115 Therefore, a direct pupillary light response should be observed
normally, but not necessarily the indirect pupillary light response.
Observation of the animal in its environment can help detect visual deficits. For
example, when new objects are introduced in their environment are they easily avoided
by the animal? Even domestic animals accommodate for any loss of sight. Often, the
animal will rely on its other senses when moving around and finding food and thus can be
difficult to assess. Polar bears, otters, and pinnipeds are more affected than cetaceans
with damage to CN II as they rely more on sight.
Cranial nerves III, IV, and VI innervate the muscles of the eyes. Cranial nerve
III, the oculomotor nerve, functions to innervate all the extrinsic muscles of the eye
except for the superior oblique and the external rectus muscles. Cranial nerve IV, the
trochlear nerve, functions to innervate the superior oblique muscle of the eye. Cranial
nerve VI, the abducens nerve, functions to innervate the external rectus and retractor
oculi muscles of the eye. Comparative studies have not been performed on these nerves
for most marine mammals. In odontocetes the nerves appear slightly reduced but the
cranial nerve function is assumed to be similar to that in other animals.116 It should be
noted that odontocetes and sirenians possess limited mobility of their eyes which can
make detecting a strabismus difficult.117 Despite that, it has been shown that dolphins are
capable of moving their eyes independently.118'119 This leads to difficulty in assessing
dolphins because if a strabismus is pathologically present, it may not be noticed. If a
strabismus is noticed, it may, in fact, be normal. Conversely, mysticetes appear to have
great mobility of their eyes.120 Dysfunction of CN III results in a ventrolateral
strabismus. Dysfunction of CN IV results in a dorsomedial strabismus. Dysfunction of
CN VI results in a medial strabismus and poor retraction of the globe on corneal reflex.103
Cranial nerve V, the trigeminal nerve, is primarily sensory to the face including
the eyelids, nasal mucosa, and cornea. It is composed of three major branches, the
ophthalmic, the maxillary, and the mandibular which also has a motor component that
innervates the muscles of the lower jaw. In most marine mammals, this is a very large
nerve (Figure 2-1).110 In mysticetes, it is the largest cranial nerve.121 This makes
functional sense in these marine mammals as large mysticetes require enormous power
for their lower jaws. In manatees and pinnipeds, this nerve is responsible for the
innervation of their tactile vibrissae on the face.122,123 Sensation of the face can be tested
by touching areas of the face such as the palpebra, cornea, and nasal mucosa.103 Dolphins
have a well developed corneal reflex.114 Lack of a facial response may be due to either
CN V or CN VII dysfunction. Loss of corneal reflex may reflect dysfunction of either
CN V or VI. With observation alone, dysfunction can be identified by the presence of
atrophy of the temporalis and masseter muscles, and a perceptible loss of sensation of the
vibrissae of sirenians and pinnipeds. Animals may have a more difficult time
manipulating their food and may present with a dropped jaw if the lesion is bilateral.103
Cranial nerve VII, the facial nerve, provides motor innervation to the muscles of
the face. In odontocetes, it is quite large due to the musculature involved in
echolocation.109 Therefore, dysfunction in odontocetes would limit echolocation and may
be associated with difficulty maneuvering and difficulty finding food. With good
eyesight and a clear tank, these signs may not be present. The same test for CN V is used
in testing CN VII, however, that requires a hands-on approach. Observable signs include
a lip, eyelid, and ear droop. Cetaceans cannot present with those signs based on their
anatomy. Pinnipeds may exhibit a lip or eyelid droop. Sirenians may exhibit a lip droop.
Otters and polar bears can exhibit all signs.
Cranial nerve VIII, the vestibulococchlear nerve, provides innervation to the inner
ear for hearing and vestibular function. In sirenians and cetaceans, ears have been
modified evolutionarily for underwater hearing and may no longer be able to detect
airborne sounds.112 Odontocetes possess extremely large vestibulococchlear nerves. This
has been hypothesized to allow the rapid and high speed conduction of acoustical
information.124 Therefore, as with cranial nerve VII, dysfunction may result in difficulty
maneuvering since they rely heavily on echolocation. In the other species, dysfunction
results in ataxia, head tilt, nystagmus, and/or deafness. Testing of this nerve in marine
mammals is difficult as it requires assessing hearing which is often quite subjective. The
absence of any vestibular signs should indicate that vestibular function is normal.
Cranial nerve IX, the glossopharyngeal nerve, controls the motor function to the
swallowing muscles. No extensive studies have been done examining this nerve in
marine mammals. Traditionally, loss of the gag reflex in animals is a sign of dysfunction.
Cetaceans do not have a gag reflex and there is no easy way to elicit this reflex safely in
the other species. Dysphagia may be the only visible sign in most animals.
Cranial nerve X, the vagus nerve, provides both sensory and motor innervation to
the swallowing muscles, as well as innervation of the heart and lungs. Here again, the
only visible sign may be a dysphagia. In the vocal animals, there may be a laryngeal
paralysis resulting in a voice change or an inspiratory stridor, if bilateral.103
Cranial nerve XI, the accessory nerve, innervates the muscles of the neck.
Dysfunction of this nerve results in associated muscle atrophy. Cetaceans and sirenians
do not possess a well defined neck area, therefore, atrophy of these muscles would not be
clearly evident. In pinnipeds, polar bears, and otters, examination of the neck muscles
should be performed through palpation in those animals that can be handled. Animals
with suspected dysfunction should be sedated for further examination.
Cranial nerve XII, the hypoglossal nerve, controls motor function to the tongue.
Dysfunction results in a loss of tongue strength or a deviation, if unilateral. Many captive
odontocetes are trained to open their mouth for examination and the tongue can be
palpated. In other animals, careful observation may be the only option.
West Nile Virus
West Nile virus, an arbovirus, is an arthropod-bome virus. The term arbovirus
includes many unrelated virus families grouped together based on their mode of vector
transmission. The arboviruses most commonly refer to members of four virus families
including Bunyaviridae, Togaviridae, Arenaviridae and Flaviviridae.
West Nile virus is a member of the family Flaviviridae. Flaviviridae is composed
of three genera: Flavivirus (from the Latinflavus meaning yellow), Pestivirus (from the
Latin pestis, meaning plague) and Hepacivirus (from the Latin hapar or hepatos,
meaning liver). 1,25,126 At one time, the flaviviruses and pestiviruses were both
considered members of the family Togaviridae, but as methods of virus identification
have broadened to include antigenic and genomic similarity, members of these groups
The flavivirus genus consists of over 70 types of related positive-stranded RNA
viruses which can be further divided into antigenically related serocomplexes.128'129 The
three main serocomplexes include the Dengue serocomplex, the Japanese encephalitis
serocomplex, and the tick-borne encephalitis serocomplex. West Nile virus is a member
of the Japanese encephalitis serocomplex which also includes Japanese encephalitis (JE),
Kunjin, Murray valley encephalitis, and St. Louis encephalitis viruses (SLEV).129 Recent
studies into the sequence homology of Kunjin has led to its proposal as a subtype of
Two major strains of WNV have been described based on genetic
relatedness.131'132 Strain I has a wide distribution encompassing Africa, Europe, and
North America while strain II has been mostly limited to Africa and Madagascar. Both
strains are detected serologically in humans and horses with strain I resulting in more
severe human and equine disease.131-133
West Nile virus is maintained in a life cycle involving birds and mosquitoes.
Infected mosquitoes feed on birds, which amplify the virus in their system. Feeding on a
viremic bird infects the mosquitoes. Once the infected mosquito has built up sufficient
viremia levels (>105 PFU/ml), it is then able to infect new hosts and the cycle repeats.
Many host species do not produce a high viremia and therefore do not have enough virus
present to infect mosquitoes.129 These species, such as humans and horses, with viremia
levels averaging 103 PFU/ml, are considered dead-end hosts.134
A host can be infected with WNV when fed upon by an infected mosquito. The
virus is released from the salivary epithelial cells of the mosquito and introduced
subcutaneously into the host.135,136 Contrary to popular belief, the virus is not deposited
in the blood because any virus deposited in the blood is reingested. Instead, the virus
replicates locally and studies have shown that the primary target for infection of the host
may be Langerhans cells.137'138 Virus replication extends into regional lymph nodes
where the virus is then carried through the lymphatics to the bloodstream via the thoracic
duct.129,139 This primary viremia seeds the extraneural tissues which act as a site of
further viral replication for release into the circulation.135'136 In unaffected hosts, the
viremia is controlled by clearance through macrophages and the appearance of circulating
antibodies which usually occurs around one week after the infection.129
In order to become infective, an arthropod vector must ingest a sufficient
concentration of virus to exceed the mesenteronal infection threshold. A vertebrate host
must have viremic titers in excess of 105 PFU/ml in order to infect a feeding mosquito.
Once in the mosquito, the virus multiplies in the mesenteronal epithelial cells. There the
virus is released systemically to further infect many tissues of the mosquito, including the
salivary glands. From the salivary glands, the virus is released and transmitted to other
vertebrate hosts during feeding.129
The mosquitoes remain chronically infected for life and produce extremely high
levels of infectious virus particles in their salivary glands.127 Although most adult
mosquitoes live for only a few weeks, some species can overwinter while infected.140 In
addition to transmission via feeding, recent studies have shown that flaviviruses can also
be vertically and horizontally transmitted between mosquitoes. Virus can infect fully
developed eggs at the time of fertilization and oviposition through the micropyle in
female mosquitoes.129'141 Venereal transmission from male to female mosquitoes has also
been demonstrated.142 In addition to mosquitoes, the virus has been found to infect hard
and soft ticks under natural and experimental conditions. The capacity of ticks to infect
new hosts, however, remains unclear.129
There is evidence to suggest that flavivirus infection can be persistent. The term
persistence has been defined by Kuno as "the prolonged presence of infectious virus,
virion components (protein, antigen and nucleic acid), and virus-specific
immunoglobulins in vertebrate hosts after infection."143 Latent infections ofWNV have
been reported in several cell cultures lines.144'145 The majority of the cells in these
cultures expressed viral antigen but a minority actually produced infectious virus.133 The
viruses that occur in these persistent infections often undergo phenotypic alterations
including a reduction in plaque size, temperature sensitivity, host-range restriction, and
loss of neurovirulence. Some cultures often exhibit alterations in the composition of their
viral proteins.133 In vivo, latent infections of WNV have been reported in ducks, pigeons,
mice, and monkeys.146-148 Whether or not there is persistence of WNV in human or
equine disease is largely undetermined.
WNV virions are small, enveloped, and spherical shaped, measuring around
50 nm in diameter.126 Virions consist of a spherical ribonucleoprotein core surrounded
by a lipoprotein envelope with small surface projections.129 The envelope is a lipid
bilayer with two or more envelope (E) proteins surrounding the nucleocapsid. The
genome of WNV consists of a positive-sense single-stranded RNA, approximately 11 kb
in length127 consisting of 199,029 nucleotides.132 The genome of WNV encodes
10 mature viral proteins.127 All the viral proteins are initially made as part of a single
long polyprotein of more than 3,000 amino acids. The polyprotein is cleaved using both
host and viral proteases. The structural proteins are encoded in the N-terminal portion of
the polyprotein with the non-structural proteins in the remainder.127 The three structural
proteins are the capsid (C), membrane (prM/M) and the E-proteins. The M-protein is a
small proteolytic fragment of the prM-protein which is involved in the maturation of the
virus into an infectious form.127 The seven nonstructural proteins are named NS 1, NS2A,
NS2B, NS3, NS4A, NS4B, and NS5.126
Studies have shown that binding and uptake may involve receptor-mediated
endocytosis via cellular receptors specific for the viral envelope proteins.127'149 Several
types of cell surface receptors have been hypothesized for flavivirus entry including
expression of highly sulfated glycosaminoglycans150'151 and antibody-dependent
enhancement.152-160 The pattern of receptor expression in animals is probably a
determinant of their tropism.127
To gain entry into the cell, the virion envelope fuses with the cellular membranes
and the nucleocapsid is delivered to the cytoplasm. Once in the cytoplasm, translation of
the genome RNA occurs using the non-structural protein NS5, which is a viral
RNA-dependent RNA polymerase (RdRp).127'149 NS5 works in conjunction with other
viral non-structural proteins and possibly cell proteins to copy complementary minus
strands from the genomic RNA template. These minus-strand RNAs, in turn, serve as
templates for the synthesis of new genomic RNAs.126,127 Viral RNA production is
asymmetric in vivo with the plus-strands accumulating more than 10 times that of the
minus strands.161'162 Virion assembly occurs within the rough ER membranes where they
bud through intracellular membranes into cytoplasmic vesicles.126'127 These vesicles
follow the host secretary pathway, fuse with the plasma membrane, and cause the release
of mature virions into the extracellular compartment.127
The primary determinant of tropism and the primary target of virus-neutralizing
activity is due to the E-protein.125 The E-protein has two major functions: it is involved
in virus attachment to susceptible cells and in the pH dependant virus-cell membrane
fusion.127,129,149,163 The E-protein is composed of three domains, of which domain III has
been proposed as a putative receptor-binding domain.125'149 Domain III contains a fold,
like an immunoglobulin constant domain, and studies have shown that altering the
Domain III structures of various flaviviruses can affect their virulence.149'151
Because the E-protein is required for entry into the cell, E-reactive antibodies are
effective at interfering and therefore, neutralizing virus infectivity.163-165 The E-protein is
responsible for eliciting high titers of virus-neutralizing, and cell membrane
fusion-blocking antibodies. However, not all E-protein-reactive antibodies neutralize
virus infectivity, which suggests that this biological activity is localized to only certain
areas or epitopes on the E-protein.163 The authentic expression of E-protein epitopes
appears to require co-expression of the prM-protein.163'166
West Nile virus was first discovered in 1937 in the blood of a woman with a mild
febrile illness in the West Nile province of Uganda.167 The first reported epidemic of
WNV was in Israel in 1951 where a total of 123 human cases were reported with no
fatalities.168 West Nile virus was subsequently isolated from humans, birds, and
mosquitoes in Egypt in the early 1950s.169,170 A serosurvey performed on humans in the
1950s in Egypt along the Nile showed that WNV antibodies were present in more than
60% of the human population.170
The first reported occurrence of equine encephalitis was in Egypt and France in
the 1960s.171,172 An outbreak in France occurred in the summer of 1962 with several
horses exhibiting neurological signs including ataxia, weakness, and amaurosis.173
Several human cases occurred in the area in subsequent years.174,175 West Nile virus was
eventually isolated from several sick horses in the area in 1965.176 In South Africa, an
outbreak in 1974 resulted in an exposure rate of nearly 55% of the human population
with clinical cases being mostly mild.177
Recently, outbreaks in humans and horses have become more frequent with
outbreaks occurring in Romania and Morocco in 1996 where nearly 500 clinical cases
were seen with a fatality rate near 10%.178-180 Other outbreaks of note have occurred in
Tunisia in 1997, Italy in 1998, Russia in 1999, and the first appearance in the U.S. in
Pathogenesis and Clinical Signs
Three main patterns of pathogenesis occur with flavivirus infections: a fatal
encephalitis with viremia, a subclinical encephalitis with a low viremia, or an inapparent
infection characterized by trace viremia and no neuroinvasion.129 The severity of the
infection on the host is dependent on several factors including the virus strain's virulence,
the level of virus transmitted and route intranasall vs. hematogenous), as well as the
ability of the host's immune system to clear the virus.129'133
The mechanism by which the virus particles cross the blood-brain barrier remains
unknown, however there are several hypotheses. One proposed pathway into the CNS
includes the transport of the virus across the cerebrovascular endothelium.185,186 The
ability of the virus to replicate in vascular endothelial cells suggests that it may grow
across capillaries.129 In fact, studies have detected infectious virus in multiple sites in the
brain and spinal cord which would support a hematogenous route.187 Another proposed
method involves gaining access to the CNS after a loss of the blood-brain barrier
integrity.18 Cytokines could play a role in the disruption of the blood-brain barrier.189
The third hypothesis involves entry through the olfactory epithelium.139,190 The olfactory
system has been previously recognized as an alternative pathway to the CNS. Olfactory
neurons are unprotected by the blood-brain barrier. In a study involving mice and
hamsters inoculated with SLEV, virus was detected in the olfactory neurons by day
4 postinoculation.190 It was further identified in the olfactory bulbs on day 5 and by day
6, it was identified in the brain. This study shows that the olfactory system may not only
be an entry way for the virus, but may result in a pathway for aerosol transmission.
Once infected, the incubation period is 1 to 6 days in humans.129 The typical case
in humans is mild, characterized by fever, weakness, headache, backache, nausea,
generalized myalgia, and anorexia.129,191,192 In European outbreaks of WNV, a roseolar
or popular rash was commonly seen, however, this has occurred less often in the U.S.193
Cases which lack a neurologic component are termed West Nile fever.192 West Nile
meningitis is defined as a meningitis with no encephalitis. West Nile encephalitis is the
most severe form and involves the brain. West Nile poliomyelitis is a flaccid paralysis
syndrome that rarely occurs with WNV infections.194
In less than 15% of patients, the symptoms are more severe and involve the
CNS.129,195 Younger patients (<50) are more likely to be affected with a meningitis while
an encephalitis occurs more often in older patients.129,192,195,196 A distinctive feature of
the North American strain of WNV is a poliomyelitis-like flaccid paralysis.129'192'195'196
In encephalitic patients, the brainstem is most severely affected and pathologically, there
are microglial nodules, perivascular cuffing, and mononuclear inflammation along the
cranial nerve roots.195,197,198 Those surviving encephalitis sometimes have residual
weakness and memory loss.195 Interstitial myocarditis and pancreatitis has also been
associated with WNV infections.129
As with humans, infections in horses can be inapparent, mild, or severe.
Experimentally about 10% of challenged horses develop clinical signs.134 Clinical signs
commonly seen in horses include ataxia, limb weakness, recumbency, and muscle
fasciculations.199 Less common signs include low grade fever (38.4 to 39.40C), paralyzed
or drooping lip, hypermetria, face or muzzle twitching, teeth grinding, hypersensitivity,
and blindness.199'200 An interstitial myocarditis has been also been associated with WNV
Lesions in horses have been characterized as a multifocal lymphocytic
polioencephalomyelitis.183 This is located in the ventral and lateral horns of the thoracic
and lumbar spinal cord and is often associated with moderate to severe hemorrhage. In
addition, there is moderate lymphocytic and monocytic inflammation with scattered foci
of microgliosis in the medulla oblongata and pons, and to a lesser extent in the basal
nuclei, thalamus, and mesencephalon. In the most affected areas there is neuronal
degeneration with central chromatolysis. There are no pathognomonic lesions in horses.
Although WNV rarely affected birds clinically in Europe, over 200 species of
birds have been reported clinically affected with WNV in the United States.192 As with
mammals, birds can experience a range of clinical signs, from asymptomatic to death and
include lethargy, recumbency, and hemorrhage in some cases.201 Emaciation, brain
hemorrhage, splenomegaly, hepatomegaly, meningoencephalitis, and myocarditis are the
most common pathological findings on clinically affected birds, however, there are not
always pathological lesions present.202'203
West Nile virus has been reported in many other wildlife and domestic species.
Clinically affected mammalian species include squirrels, bats, chipmunks, rabbits,
raccoons, dogs, and cats.8'204-207 There have been reports of several canids with clinical
WNV infection.9 A wolf pup and an adult 8 year old dog exhibited clinical signs
including anorexia, weakness, ataxia, ptyalism, head tilt, and blindness.9 The dog had a
concurrent immune-mediated thrombocytopenia which might have made him more
susceptible to disease. The encephalitis affected the gray matter with poorly demarcated
aggregates of lymphocytes and microglial cells and mild necrosis. Marked myocarditis
was also present.
Many more mammals have seroconverted to WNV without signs of clinical
disease. While seroconversions in marine mammals have been reported, captive P.
vitulina and M. schauinslandi have also been clinically affected.3'5'6 These seals
exhibited clinical signs including weakness, head tremors, anorexia, and abnormal
swimming. Lesions included a nonsuppurative encephalitis, meningitis and myocarditis.
Reptile and amphibians are also susceptible to WNV. Experimentally infected
lake frogs (Rana ridibunda) not only developed a high titer, but were able to transmit the
virus to Culexpipiens.208 More recently, WNV infected captive alligators have exhibited
clinical signs such as "star gazing", ataxia and muscle spasms.
As of 2004, WNV has been detected in all of the continental U.S. and has
expanded into Canada, Mexico and several Caribbean islands.209-211 West Nile virus will
likely continue spreading through the remainder of North America, and into Central and
Mosquitoes are involved in an enzootic cycle, transmitting the virus to additional
hosts. The main mosquito species involved in maintaining the bird/mosquito cycle are
Cx. restuans and Cx. pipiens.212 The high prevalence of virus detected in these
ornithophilic vectors early in its introduction helped WNV spread rapidly through the
U.S. via the bird populations. The mosquito species of greatest interest to human
research are those who feed on both birds and mammals, as they provide the link between
the viremic birds and mammalian hosts. As of 2004, WNV had been isolated from
43 different species of mosquitoes.213 In the U.S., the majority of isolates have come
from the Culex species, primarily Cx. restuans, Cx. pipiens, and Cx. salinarius as well
as some Aedes species.214'215 As WNV spread southward, additional species such as Cx.
nigripalpus and Cx. quinquefasciatus became involved in the transmission cycle. Cx.
salinarius appears to be mammalophilic compared to Cx. pipiens and Cx. restuans and as
such, may be an effective bridge from birds to mammals.192 In the western states the
species Cx. tarsalis is an indiscriminate feeder of both mammals and birds, and may be
the reason why western states such as Colorado had such high infection rates in
2003.192,216 Culex is also an efficient vector because it can overwinter as an adult
The first report of WNV in humans in the U.S. was in August of 1999.217 The
introduction of WNV became evident as several clinical cases of encephalitis in humans
were reported at the same time that unexplained mortality of birds, primarily crows, was
occurring 1 Testing of the samples using RT-PCR showed >99% similarity with a strain
from Israel.132'163'217 The mechanism by which the virus was introduced into the U.S. is
not known. Hypotheses include introduction of an infected bird or mosquito.
Since its introduction, human cases have occurred in nearly every state in the
continental U.S., moving in a westerly direction.218 Most infections have occurred via
mosquito transmission and as such, increased exposure to mosquitoes directly increases
the chances of becoming infected. Therefore, the amount of time spent outdoors, as well
as the season, can affect ones chances of infection. In some areas there is seasonal
blood-feeding of the mosquitoes and this results in higher infection rates during certain
time periods.192 However, as WNV moves into tropical areas, the mosquitoes feed year
round and there will likely be no seasonality.
Once infected, the determination of the pathogenesis is a result of host factors
including age, sex, genetic susceptibility, and pre-existing immunity to heterologous
agents.129 Immunosuppression and age (>60) have both been hypothesized to increase
the chances of developing clinical signs.129,219 In addition, older patients were more
likely to die from the disease as compared to younger patients.219
Several other modes of WNV transmission in humans have occurred in the U.S.
There are reports of WNV transmission from the mother both transplacentally (1 reported
case),220 and via breast milk (1 reported case).221'222 In addition, WNV clinical cases
have occurred after blood transfusions from infected individuals (4 reported cases) and
organ transplants (4 reported cases).222-225 Transmission has also occurred through
exposure in the work place at a turkey farm and after accidental inoculation in a
Birds amplify WNV as part the virus' life cycle. In 1997, a new strain of WNV
was identified clinically affecting birds in Israel.228 Previously, WNV was rarely
associated with clinical disease in birds. As previously mentioned, the strain of WNV
that arrived in North America was nearly identical to the strain from Israel.132'163'217 This
strain has resulted in significant avian mortalities in North America.
Thus far, WNV has caused fatal infections in 198 species of birds.192 Incidence in
certain species, such as the American crow (Corvus brachyrhynchos), has been extremely
high, with some areas approaching near 100% mortality.192,203 A closely related species,
the blue jay (Cyanocitta cristata), has also been severely affected.192 Studies performed
in New York during the outbreaks in 1999 and 2000 found that nearly one third of the
avian species tested had virus-neutralizing antibodies to WNV.203,229,230
To be an effective reservoir for WNV, a bird species must be frequently exposed
to infection, abundant in relation to other bird species, and capable of producing viremic
levels high enough to infect mosquitoes.229 Several bird species develop extremely high
levels of viremia and are very competent amplifying hosts.201 The Passerine species such
as the blue jay, common grackle, house finch, house sparrow, and American crow all are
competent at transmitting virus back to mosquitoes as are the Charadriiformes such as
gulls, terns and plovers. Because of their high abundance in many areas, house sparrows
are considered an important reservoir host even though their seroprevalence is not
generally as high as crows.229 In contrast, the Piciformes (woodpeckers, toucans),
Psittaciformes (parrots), and Galliformes (chickens, turkeys) have all been identified as
incompetent vectors.201 These birds are fairly resistant to disease and do not develop
high viremic levels.
Early on in the outbreaks it was clear that dead birds provided early warning
surveillance for WNV. In 2000, dead bird surveillance was initiated in New York to
provide a method of detecting early virus activity before the onset of human cases.203
The best and earliest warnings were provided by dead crows. Although studies have
shown that the presence of WNV positive birds does translate to local virus transmission,
this does not indicate the levels of viral transmission in the area.212 In order to detect
local virus transmission, several species were considered as possible sentinels. The ideal
sentinel candidate would seroconvert to WNV upon exposure, but not show any clinical
signs or develop a viremia high enough to infect mosquitoes. Experimental infections in
goslings, poults, and chickens have been performed to evaluate their use as sentinel
species.202'231-233 Results showed that chickens shed low amounts of WNV per cloaca and
per os and that they developed low level viremias.231'233 In the case of the goslings and
poults, both produced viremias higher than the chickens. The goslings were clinically
affected.202'232 In addition there was low level cloacal shedding in the poults and low
level oral shedding in the goslings. Because of this, chickens have been viewed as safe to
use as sentinel species.
Oral and cloacal shedding of virus has been demonstrated in many bird species,
however, virus levels are generally low.192 Oral transmission via infected water, dead
birds, mice, and mosquitoes has been shown experimentally.201 Horizontal transmission
among birds confined together has been demonstrated in several species.192,201,202 This is
important because despite effective control of mosquito vectors, the virus may continue
its life cycle in this manner in some areas.
Significant morbidity and mortality has been reported in equids in several
outbreaks in Europe and the Middle East.171,234,235 A WNV outbreak in France in
2000 resulted in 76 equine clinical cases with 21 deaths.171'236 As a results, the local
animal health authority performed a serosurvey on all equids located within a 10 km
radius of the confirmed cases.236 Results on 1,429 horses showed a prevalence of IgG
antibodies of 19.2% (274 horses). The first confirmed case of WNV infection in a horse
in the Western Hemisphere was in October of 1999.199 In 2000, 60 horses were
confirmed as having a case of WNV encephalitis with a fatality rate of 38%, either due to
natural causes or euthanasia. With each year, the number of equine cases has grown
exponentially.192 This may be due in part to the movement of WNV into areas where the
Culex vectors feed more frequently on horses.
Horses are considered dead-end hosts because they produce a maximum viremia
of 103 PFU/ml, which is well below the 105 PFU/ml threshold to infect mosquitoes.134
Studies in horses have failed to show any correlation between age or gender and infection
rates in horses.192 The primary risk factors for horses involves their proximity to
communal bird roosts or waterfowl congregations.
Beginning in 1999, serum samples were taken from dogs and cats in New York to
determine their exposure to WNV.237 Antibodies were detected in 5% of the dogs and in
0% of the cats, and none demonstrated clinical signs. Similar studies have been
performed in Europe to examine the prevalence of WNV antibodies in dogs using the
hemagglutination test.238 There was a reported seroprevalence of only 0.7%, however,
these results could not be completely interpreted because tick-borne encephalitis is also
present in the area and is cross-reactive.
Recent studies have evaluated experimental infection of dogs and cats with WNV
to determine their response to infection as well as their potential to serve as amplifying
hosts.205 The dogs were exposed to WNV through the experimentally caused bites of
infected mosquitoes. None of the dogs exposed developed any clinical signs, nor
developed high levels of viremia. Therefore, dogs are considered dead-end hosts.
Cats were exposed to WNV through the experimentally caused bites of infected
mosquitoes, while four additional cats were exposed via oral ingestion of mice previously
infected with WNV through an intraperitoneal inoculation.205 All of the cats tested
developed higher viremia levels than the dogs. Of the four cats inoculated via
mosquitoes, three of the four exhibited lethargy and a febrile response, but no
neurological signs. The four cats who ingested WNV infected mice developed similar
levels ofviremia but clinical signs were absent. This was the first time transmission via
ingestion was reported. Cats have high enough virus levels to infect mosquitoes,
although it is probably much less efficacious when compared to many avian hosts.
Studies into the disease status in most exotic and wildlife species are lacking.
Recent serosurveys have been performed on captive primates in Louisiana, including
rhesus macaques, pigtail macaques, and baboons.239 Results indicated that approximately
36% of the primates had been exposed to WNV, indicated by PRNT testing. WNV has
also recently been found to affect alligators.7'240 The manner in which the alligators were
originally exposed is still unclear. The alligators were farmed and had been fed
horsemeat that tested positive via PCR for WNV. This has been suggested as a means of
transmission, but has not been proven. The alligators were also under stressful growth
conditions including year-round warmed-water and being housed in complete darkness,
which may have resulted in immunosuppression. Alligators may prove an important
reservoir host as they develop high viremia levels and may harbor virus overwinter.240 In
addition, lateral transmission and transmission via ingestion has also been shown
Diagnosis can be difficult because WNV resembles other related viruses and there
are no pathognomonic signs for WNV. Most hematological and biochemical parameters
remain within normal limits. Cerebrospinal fluid findings include mononuclear
lymphocytic pleocytosis and elevated protein.195
The viremia in most mammals is relatively low and of short duration, so virus
isolation is rarely performed. Growth of virus in cell culture remains the gold standard
for viral detection but requires time to grow and to be identified.241 Flaviviruses can be
cultured in whole animals such as chick embryos, suckling mouse brain, and mosquitoes,
as well as in primary or established cell lines of mammalian, avian, or insect origin.127
Serological testing is based on detection of IgM and/or neutralizing antibodies
after exposure to WNV. Strongly neutralizing antibody is the most virus-specific.163 In
addition, IgM antibodies are more specific than IgG antibodies.242'243 The PRNT is the
gold standard for the serological diagnosis of WNV infection. This test has the
advantage of differentiating between WNV and closely related flaviviruses.244 The
disadvantage of this test is that it requires the use of live virus in biosafety level
3 containment and is time consuming.245 The MAC ELISA is highly specific and
measures serum IgM antibody levels. These levels rise and fall rapidly in the horse
compared to neutralizing responses to WNV.199 A positive MAC ELISA is interpreted as
exposure to WNV within 30 days. IgM antibodies are detectable around day
4 post-infection which usually coincides with the first signs of clinical disease.133,244
Neutralizing antibodies generally appear around 4 to 5 days after the onset of
illness.133'243 In the CSF, the switch to IgG appears to be earlier, occurring around day
3 to 4 of illness. In humans, IgM can occur in the CSF before detection in the serum.
Because this occurs, it is assumed that IgM production occurs locally due to intrathecal
viral infection and not from the systemic circulation. Neutralizing antibodies are
detectable for several months to years after acute infection. Over 90% of convalescent
sera are positive 2 and 3 months after the onset of illness.243
In addition to the MAC ELISA, blocking or competitive inhibition (CI) ELISA
testing has been performed for several different flaviviruses using monoclonal
antibodies.246-249 Studies using monoclonal antibodies, namely 5H10, 3A3, 7H2 and 5C5,
have shown that all monoclonals strongly neutralized WNV strain I but not strain II.125
These experiments have shown that each monoclonal recognizes spatially distinct epitope
RT-PCR has the advantage of speed, specificity, and sensitivity for the detection
of viral RNA, but its utility is also dependent on the viremia.241'250 Although the
sensitivity and specificity of the assay are excellent, the many steps involved and the
specialized lab necessary for the work made it an inefficient method to detect WNV for
large numbers of birds.251 The VecTest WNV/Saint Louis encephalitis virus Antigen
Panel Assay (Medical Analysis Systems, Inc., Camarillo, CA) was originally developed
to detect WNV or SLEV in infected mosquitoes.252 However, recent studies have used
this assay to detect the presence of WNV in oropharyngeal and cloacal swabs taken from
birds.251'252 Results indicated that oropharyngeal swabs were more sensitive than cloacal
swabs when used for antigen detection.25 In American crows, the sensitivity of the test
was 83.3% with the specificity being 95.8%, making the assay an acceptable screening
test for field use. The main advantage of this test is its speed, with results available
within 15 minutes. The main disadvantages are that it has a lower sensitivity when
compared to the RT-PCR TaqMan assay, as well as a decreased efficacy when working
with non-corvid species.
Treatment and Prevention
There is currently no treatment for WNV infection and treatment is primarily
supportive. In vitro studies have found antivirals such as IFN-a2b to be effective against
the virus while ribavirin was protective but not therapeutic. 191,253,254 In vivo studies using
mouse and hamster models has shown promise with IFN-a2b therapy, while ribavirin
treatment has been unsuccessful.255'256 Recent studies with Dengue virus infections have
shown that mycophenolic acid, a non-nucleoside inhibitor of IMP dehydrogenase, has
inhibited Dengue virus infection by reducing the levels of RNA produced.257
Epidemiological studies support a point source exposure to WNV, thus
vaccination is a primary focus of prophylaxis. A formalin-inactivated vaccine has been
marketed since 2001 for horses. Both horses and birds have been vaccinated using this
product. Inactivated vaccines have the advantage of being relatively safe for animals and
are relatively simple to develop. The main disadvantage of these vaccines is that they
generally require multiple inoculations to produce an effective response. These vaccines
generally do not elicit a sustained cellular immune response.258
Live attenuated vaccines should elicit a much stronger humoral and cellular
immune response and have the advantage of requiring only a single dose. A recombinant
vaccine against WNV using yellow fever 17D as a live vector for the envelope genes of
WNV and other flaviviruses has been developed and is presently in human safety trials.
Similar vaccines for Japanese encephalitis and dengue are currently under development
and testing.258,259 Additional vaccine candidates include recombinant DNA vaccines
expressing various proteins and recombinant E-protein subunit vaccines.260-262 The final
option to prevent WNV infections is through mosquito control.
DEVELOPMENT OF A COMPETITIVE INHIBITION ENZYME-LINKED
West Nile virus (WNV) is an arthropod-borne virus introduced to the New World
in 1999 that causes neurological disease primarily in birds, horses, and humans.1 Since
its initial appearance, this virus has spread rapidly across the North American continent.3
During 2003, WNV was established in much of the U.S., affecting over 9000 people with
WNV encephalitis and WNV fever. Thus far, harbor seals, monk seals, alligators, gray
squirrels, alpacas, llamas, a wolf, and one dog have been clinically affected by natural
infections.4,12,13,17'18'22'23'28 A wide range of other animals have seroconverted to WNV,
although actual clinical status in these instances has not been documented. The infection
and disease status has not been completely assessed for many species because of limited
federal and state resources. Since WNV has been detected in a variety of birds and
mammals, assessment of multiple species of free-ranging and captive wildlife species is
necessary to assess the risk of WNV infection to these populations. A survey of WNV
status in non-domestic animals in the state of Florida should include St. Louis
encephalitis virus (SLEV) since this closely related Flavivirus cross-reacts with WNV
antigens and could confound testing.21
The tests of choice, allowing highly specific serologic testing for viruses, are
neutralization tests. These tests are applicable to multiple species, but are expensive as
screening tools, requiring viral expertise, special facilities, and often special permits for
zoonotic agents. The goal of this research was to develop and validate a WNV specific
ELISA using monoclonal antibodies in a competitive inhibition format (CI ELISA).
Currently, there are several serological methods utilized to identify WNV virus
exposure in animals. The hemagglutination-inhibition test (HIT) is a commonly used
screening test for WNV. These assays are labor intensive and require the use of
nonspecific inhibitors. In addition, since related flaviviruses share many of the same
epitopes, there is a high degree of cross-reactivity.26 While the reverse
transcription-polymerase chain reaction (RT-PCR) is an extremely sensitive and specific
method of detecting WNV antigen in tissues, it has limited value with serum or
cerebrospinal fluid (CSF).14 Since the viremic stage occurs before the onset of clinical
signs in mammalian hosts, virus is no longer present in plasma once clinical signs occur.
Immunoglobulin M antibody capture ELISAs (MAC ELISA) and IgG direct ELISAs are
both utilized.24 The MAC ELISA has a high degree of sensitivity and specificity for both
CSF and serum, but requires the use of species-specific antibodies. The IgG ELISA also
requires the use of species-specific antibodies and has a higher degree of cross-reactivity
than the MAC ELISA. Various monoclonals have been tested in their ability to react in a
CI ELISA format for agreement with the PRNT in domestic mammals and birds.6'7
Utilizing a panel of monoclonals and several antigens, this research tested whether or not
the reagents could be used in a CI ELISA as a rapid WNV screening assay with broad
applicability to multiple wildlife species.
Materials and Methods
Serum samples from marine mammals were obtained from various U.S. zoological
institutions. All blood collection was taken for routine purposes by the institutions or
researchers, and not for the purpose of this study. Sites included: Miami seaquarium,
Columbus zoo and aquarium, Lowry Park Zoo, Rio Grande zoo and Six Flags Ohio. In
addition, samples from wild manatee captures were provided on a yearly basis by the
Florida Fish and Wildlife Conservation Commission's Fish & Wildlife Research Institute
(FWRI), and by the United States Geological Survey (USGS) via their Sirenia project.
Samples were tested at the University of Florida, College of Veterinary Medicine in
Gainesville, Florida. It was conducted under authorization of the IACUC, number
C411, the United States Fish and Wildlife (USFW) permit numbers MA791721-3 and
MA067116-0 and the National Oceanic and Atmospheric Administration (NOAA) permit
In addition to marine mammals, several other species of animals were tested. In
conjunction with the Ringling Brother's circus, elephant serum samples obtained for
routine blood work were provided from several populations of elephants in the state of
Florida. Alligator serum samples were provided by Dr. Elliot Jacobson. Archived horse
and domestic cat serum was also used. The remaining serum samples were from various
exotic species collected for routine clinicopathological testing and were provided by the
Clinical Pathology department at the University of Florida's Veterinary Teaching
Hospital from the years 2002 and 2003.
Two sources of WNV antigen were utilized for the CI ELISA and included WNV
mouse-brain antigen from Bioreliance (Rockville, MD) and inactivated WNV
recombinant antigen developed by Acambis Inc. (Cambridge, MA gifted by Intervet
Incorporated). Positive and negative controls consisted of equine serum obtained from
NVSL (Ames, IA). The monoclonal antibodies, 7H2, 3A3, and 5H10, were purchased
from Bioreliance (Rockville, MD). Horseradish peroxidase (HRP)-conjugated goat
anti-mouse IgG antibody was purchased from Amersham Biosciences (Piscataway, NJ).
The substrate used was 3, 3', 5, 5'-tetramethylbenzidine solution (TMB) (KPL,
Plaque Reduction Neutralization Test
A total of 262 serum samples were tested representing 28 different species of
animals. Samples were kept frozen and shipped overnight to the Florida Department of
Health where they were tested using the plaque reduction neutralization test (PRNT).
The samples consisted of 212 different individual animals. Several animals were
sampled more than once.
Briefly, samples were titered out from 1:10 and 1:100 for testing in a sample buffer
composed of M-199 salts, 1% bovine serum albumin, 350 mg/L sodium bicarbonate,
100 units/mL penicillin, and 100 mg/L streptomycin in 0.05 Tris, pH 7.6 in sterile
microtiter plates. Virus (75 mls-100 PFU) and 75 mL of serum was mixed and incubated
at 37C for 75 minutes. After incubation, the mixture was added to flasks containing
confluent monolayers of Vero cells. Following incubation at 370C for 60 minutes, flasks
were overlaid with agar and incubated for 72 hours. A second agar overlay containing
neutral red was added and the flasks examined after 24 hours. Plaque reduction of
>90% was considered positive. A PRNT titer of at least 1:10 in serum was considered
positive and specific for WNV.
Competitive Inhibition Enzyme-Linked Immunosorbent Assay
A 96-well Immulon 2HB plate (Fisher, Pittsburgh, PA) was coated with
mouse-brain antigen at a dilution of 1:100 in coating buffer (carbonate/bicarbonate pH
9.6), and allowed to incubate overnight at 40C. The plate was washed and the wells
blocked with 150 p.L 5% Blotto (5% instant powdered milk in 20 mM Tris pH 7.6, 0.15
NaC1, 0.05% Tween20), sealed and incubated for one hour at room temperature. The
serum/plasma samples were diluted 1:2 in the serum/control diluent and 60 ptL placed on
a transfer plate (Fisher, Pittsburgh, PA).
Monoclonal antibodies were diluted at 1:1000 (3A3 and 5H10) and 1:5000 (7H2)
in the serum/control diluent. These dilutions were based on individual reactivity against
mouse-brain WNV positive and negative antigen (data not shown). Sixty itL of the
monoclonal (MAb) mixture was added to each well of the transfer plate, except for the
blank and the non-specific binding (NSB) wells. The MAb wells contained the
serum/control diluent and the monoclonal antibody mixture only. The NSB and blank
wells contained the serum/control diluent only. All other wells contained the sample and
the monoclonal antibody mixture. A serum/control diluent was prepared consisting of
48 mL of phosphate buffered saline with Tween-20 (PBST) and 2 mL of normal goat
serum. A conjugate diluent was prepared consisting of 48 ml of 5% Blotto and 2 ml of
normal goat serum. Following blocking, the Immulon 2HB plate was washed and
100 p.L of the sample/monoclonal combination was transferred to the corresponding
wells of the Immulon 2HB plate. The plate was incubated for one hour at 370C.
Following incubation, the plate was washed and 100 ptL of HRP-conjugated goat
anti-mouse IgG antibody diluted at 1:1000 in conjugate diluent was added to each well
except the blank, which contained PBST only. The plate was incubated for 30 minutes at
37C. The plate was washed and 100 p.L of TMB substrate was added to each well. The
plate was allowed to develop for 15 minutes at room temperature in a dark drawer. The
reaction was stopped after 15 minutes with 100 ptL of 1% HC1 solution and the plate read
at 450 nm using a Bio-Tek plate reader (Bio-Tek Instruments, Winooski, VT). Positive
samples were determined by measuring the percent inhibition of the negative control
using the optical density (OD). The formula used for this was [100-(sample OD/negative
control OD* 100)].
In order to determine whether or not a combination of three monoclonal antibodies
was superior to running a single monoclonal, plates were tested using varying dilutions of
the monoclonal antibodies. Plates were coated using the WNV positive mouse-brain
antigen at 1:100 and the CI ELISA was performed as previously described. Each
monoclonal was run both separately and as a group to compare results. Positive,
negative, blank, and NSB controls were run on all plates. All monoclonals were
compared against each other and against the monoclonal combination. The best
performing monoclonal preparation was then serially diluted to determine optimal
dilution on a checkerboard plate.
Antigen and monoclonal optimization
Separate checkerboard plates were tested to determine the optimal dilution of two
separate antigens in combination with several dilutions of the monoclonal antibody 7H2.
One plate was coated with two dilutions of the mouse-brain antigen, 1:50 and 1:100.
Another plate was coated with inactivated West Nile ChimeriVax antigen (Intervet,
Millsboro, DE) in a checkerboard fashion with dilutions including 1:2, 1:4, 1:8, 1:16,
1:32, 1:64 and 1:128. For both antigens, different dilutions of the 7H2 monoclonal were
tested including 1:2000, 1:4000, and 1:5000. Positive, negative, blank, and NSB controls
were run on all plates.
Available samples that had been tested using the PRNT were aliquotted, blinded,
and randomized for testing with the CI ELISA. After the completion of testing of all the
samples, the sample numbers were matched up with previous PRNT test results to
determine the proper cut-off value to be used for the CI ELISA based on statistical
analysis. One plate was selected at random and run in triplicate on different days to
determine inter-assay variability. Sensitivity and specificity values were performed by the
following formulas: Sensitivity was calculated by using the formula [True positives/
(True positives + false negatives)]. Specificity was calculated by using the formula [True
negatives/ (True negatives + false positives)]. Statistics were calculated using a
commercial statistics software program (MedCalc, Mariakerke, Belgium). The
coefficient of variance was calculated by the formula (standard deviation/average) x 100.
There were 262 samples from 28 species collected and tested by PRNT for the
presence of neutralizing antibody against WNV and SLE (Table 3-1). Seven species
(12 individuals) were positive for WNV neutralizing antibody and these included alpacas,
camels, cougars, elephants, lions, tigers, and a box turtle. Four species (11 individuals)
were positive against SLEV and these included camels, elephants, a lion and a tiger.
When positive, WNV titers ranged between 10 and 640 and SLEV titers ranged between
20 and 640 (Table 3-1). 3 species (8 individuals) had titer values for both WNV and
Table 3-1. Plaque reduction neutralization results by species
Collection dates Species # WNV/ # SLE/ WNV titer
#Tested #Tested SLE
Oct 2002 Alpaca 2/3 0/3 >320;>40
May 2002 Bat 0/1 Not tested Not applicable
May-Aug, Oct 2002 Bovine 0/10 0/10 Not applicable
Dec 2002 Camel 2/2 2/2 1:640; >320
Jun 2002 Capuchin 0/1 0/1 Not applicable
Nov 2002 Cougar 1/2 0/2 1:320
Jun, Sep 2002 Deer 0/3 0/2 Not applicable
Mar, May, Jul, Aug, Bottlenose 0/12 0/12 Not applicable
Oct, Dec 2002 dolphin
Jan, Apr 2003 Asian 4/26 7/26 1:10; >40; 1:80;
1:10; 1:10 1:20;
1:40; >40; 1:80;
Jun 2002 Ferret 0/1 0/1 Not applicable
May, Jun, Aug-Dec Goat 0/12 0/5 Not applicable
Nov 2002 Kangaroo 0/1 0/1 Not applicable
May, Aug, Killer 0/3 0/3 Not applicable
Oct 2002 whale
May 2002 Leopard 0/1 Not tested Not applicable
Jun, Aug 2002 Lion 1/2 1/1 >320
Aug, Sep 2002 Llama 0/5 0/1 Not applicable
Jul 2002 Macaque 0/1 Not tested Not applicable
Jan-Apr, Jun-Sep, Manatee 0/136 0/69 Not applicable
Dec 2002 Monkey 0/1 0/1 Not applicable
Jan, Feb 2003 Monk 0/21 0/21 Not applicable
Jul, Aug 2002 Porcupine 0/2 Not tested Not applicable
Sep, Nov 2002 Rabbit 0/2 0/1 Not applicable
Mar, Apr, Jul, Dec Sea lion 0/4 0/4 Not applicable
Aug 2002 Sheep 0/1 0/1 Not applicable
Jun 2002 Spider 0/1 0/1 Not applicable
Jan, May, Jun, Dec Tiger 1/4 1/4 1:160
Aug 2003 Box turtle 1/1 0/1 1:10
Aug 2003 Tortoise 0/1 0/1 Not applicable
The highest optical densities were achieved with monoclonal 7H2 alone or by all
three monoclonals used simultaneously. The optical densities for 7H2 MAb alone was
nearly identical to that obtained by using all three MAbs (Figure 3-1).
0.3 -- Mab
0\ 0 5 b l %b N ,.
Figure 3-1. Comparison using MAbs 3A3, 5H10, and 7H2 separately and as a group
(Mab). Plates were coated with mouse-brain derived antigen at a 1:100
dilution. Monoclonal dilutions were 1:1000 (3A3, 5H10) and 1:5000 (7H2).
0D 6 -Ag 1:10
at 1:2000 at 1:3000 at 1:4000 at 1:5000 at 1:6000 at 1:7000
Figure 3-2. Determining optimal 7H2 dilution by testing several dilutions on the CI
ELISA. Plates were coated with mouse-brain antigen at a 1:100 dilution.
Because there was no additional contribution of 3A3 and 5H10, serial dilutions of 7H2
were investigated. Results indicated that a 7H2 dilution of 1:4000 was appropriate
Antigen and Monoclonal Antibody Optimization
A total of nine plates were run using the mouse-brain antigen and compared to each
other for inter-assay variation. Results using the mouse-brain antigen were inconsistent
with an inter-assay variation based on a calculated coefficient of variance (CV) of
21.8% with a standard deviation of 7%. The NSB using the mouse-brain antigen was
extremely high, ranging from 0.08 to 0.33. The two dilutions of mouse-brain antigen
exhibited little difference. The dilution of 1:50 resulted in a slightly higher optical
density when compared to the 1:100 dilution (Figure 3-3).
OD0.4 --Ag 1:50
0.3 ----Ag 1:100
at 1:2000 at 1:4000 at 1:5000
7H2 monoclonal dilutions
Figure 3-3. Plates were coated with mouse-brain antigen at two separate dilutions
(1:50, 1:100). Several dilutions of the 7H2 monoclonal were tested against
the antigen dilutions for reactivity.
The 7H2 performed similarly at the dilutions 1:2000 and 1:4000, but levels dropped off at
1:5000. Using the ChimeriVax antigen at dilutions 1:2 through 1:32 resulted in optical
densities that were nearly identical for all controls (Figure 3-4). The dilution of
1:64 resulted in the highest optical density with good separation between positive,
negative, and monoclonal controls. The 1:128 dilution was slightly higher, but still not at
the level of 1:64. Inter-assay variation based on a calculated coefficient of variance (CV)
for three plates consisting of 35 samples was 7.9% with a standard deviation of 4.9%.
OD -1- Positive
0.8 -- Negative
at 1:2 at 1:4 at 1:8 at 1:16 at 1:32 at 1:64 at 1:128
Figure 3-4. ChimeriVax antigen was serially diluted and used to coat a plate. Positive,
negative, and 7H2 controls were tested against each dilution for reactivity.
The cut-off for interpretation of the CI ELISA results was determined by two
different values, raw optical density and percent inhibition using only the horse and
alligator samples. The first method was based on the OD compared to the results of the
PRNT. Utilizing a receiver operating characteristic (ROC) curve analysis, the OD cut-off
value was determined to be at 0.550 with a sensitivity of 92.9% with a 95% confidence
interval of (76.5 to 98.9) and a specificity of 97.2% and a 95% confidence interval of
(90.2 to 99.6). All animals above the 0.550 cut-off would be considered negative for
WNV exposure (Figures 3-5 and 3-6).
0 20 40 60 80 100
Figure 3-5. An ROC curve demonstrating the sensitivity and specificity of the CI-ELISA
using an OD cut-off value of 0.550.
Figure 3-6. Scattergram demonstrating the separation of the positive WNV animals (1)
and the negative WNV animals (0).
Using an ROC curve analysis, the percent inhibition cut-off was determined to be
at 28% inhibition. Any sample testing greater than 28% inhibition would be considered
WNV positive. Using this cut-off there was a sensitivity of 92.6% with a
95% confidence interval of (75.7 to 98.9) and a specificity of 100.0% with a confidence
interval of (94.9 to 100.0) (Figures 3-7 and 3-8).
0 20 40 60 80 100
Figure 3-7. An ROC curve demonstrating the 100% specificity and 92.6% sensitivity
with a large area under the curve using a percent inhibition cut-off value.
Figure 3-8. Scattergram demonstrating the separation of the positive animals (1) and the
negative animals (0) using the 28% inhibition cut-off value.
The sensitivity and specificity were also analyzed by separate species. Horse
samples had a sensitivity of 84.6% and a specificity of 96.2% using an OD cut-off of
0.549. Their distribution is demonstrated in Figure 3-9. When examining the alligator
data alone, there is 100% specificity and sensitivity using a 28% inhibition cut-off value
(Figure 3-10). Domestic cats were tested using the CI ELISA; however, none of the
available samples were positive by either WNV or SLEV PRNT. A small number of
exotic cats were also examined. There was 100% agreement between the CI ELISA
interpretation and the PRNT results using the predetermined 28% cut-off for both the
exotic and domestic cats (Table 3-2).
** *. *E *
Figure 3-9. Distribution of PRNT-tested positive and negative horses tested on the CI
ELISA and compared to a 28% inhibition cut-off value.
Alligator plasma samples
Figure 3-10. Scattergram representing results of known positive and negative alligator
plasma samples as tested on the CI ELISA.
Table 3-2. Agreement between cat samples on the CI ELISA and the PRNT. All
samples are domestic cat unless otherwise noted.
Animal # Date sampled WNV PRNT titer Percent inhibition of
4697 2/27/04 <10 0/0.9255
4715 2/27/04 <10 0/0.756
4717 2/27/04 <10 0/1.0835
4735 2/27/04 <10 13.8/0.647
4795 2/27/04 <10 0.03/0.7485
93192 8/7/02 <10 0/1.067
96191 7/23/02 <10 0/0.9805
103160 9/4/02 <10 0/1.26
130393 8/21/02 <10 0/0.9505
152563 8/7/02 <10 0/0.9575
152563 8/14/02 <10 0/0.9215
158542 7/19/02 <10 0/0.718
158721 8/27/02 <10 0/1.1495
160137 8/15/02 <10 0/0.869
160660 7/16/02 <10 0/0.9625
160739 8/27/02 <10 0/1.2825
161429 8/14/02 <10 0/1.159
164597 7/29/02 <10 0/1.1565
164714 8/12/02 <10 0/1.019
165326 8/15/02 <10 0/1.975
165356 7/29/02 <10 0/1.304
165399 7/24/02 <10 0/0.85
165408 7/29/02 <10 0/1.132
Table 3-2. Continued.
Animal # Date sampled WNV PRNT titer Percent inhibition of
165476 8/5/02 <10 0/1.1125
165750 7/30/02 <10 0/0.93
165804 8/12/02 <10 0/1.105
165908 8/14/02 <10 0/1.0435
165927 8/14/02 <10 0/1.019
166186 9/4/02 <10 0/1.043
166566 8/12/02 <10 0/0.9905
A1B5 2/27/04 <10 8.4/0.6875
A1C7 2/27/04 <10 3.2/0.7265
A1F7 2/27/04 <10 20.9/0.594
AGD6 2/27/04 <10 11.1/0.7345
AGK5 2/27/04 <10 14.1/0.71
AHD2 2/27/04 <10 19.6/0.6475
AHS5 2/27/04 <10 14.4/0.6425
AHT4 2/27/04 <10 10.7/0.67
AIC6 2/27/04 <10 24.5/0.591
AICY 2/27/04 <10 13.9/0.712
IFH3 2/27/04 <10 0/1.0635
IGK4 2/27/04 <10 0/0.8585
IGN2 2/27/04 <10 18.9/0.6085
IGW2 2/27/04 <10 3.1/0.801
Q1G2 2/27/04 <10 16.2/0.693
QDCI 2/27/04 <10 0/1.4825
QGS6 2/27/04 <10 0/0.9665
QHV4 2/27/04 <10 22.0/0.5855
QJA5 2/27/04 <10 13.7/0.648
QJG1 2/27/04 <10 19.8/0.602
163563 (leopard) 5/3/02 <10 0/1.047
163099 (tiger) 5/8/02 <10 0/0.9195
167804 (cougar) 11/6/02 <10 0/0.978
144566 (tiger) 1/13/03 >40 38.7/0.507
144566 (tiger) 1/10/03 1:160 54.9/0.353
Of the 20 non-manatee marine mammal samples tested, only one was PRNT
positive. This particular animal was a clinically ill harbor seal (gift from Dr. Ned Gentz
at the Rio Grande Zoo). Using the 28% cut-off, the distribution of the samples is shown
in Figure 3-11. Of the 126 manatees tested, all tested negative on the PRNT assay.
Figure 3-12 illustrates the manatees that tested above the 28% cut-off using the
CI ELISA. There are four false positives.
Marine mammal samples
Figure 3-11. Scattergram of the non-manatee marine mammal samples as tested on the
CI ELISA using the percent inhibition cut-off.
II 1 "
' I I |
I'I 1 1I
Level above a 28% cut-off for each manatee tested. Note that any bars
going above the zero line would be suspect positives requiring further
There is disagreement of the positive and negative elephants (Table 3-3) based on
their performance on the CI ELISA. Using a 28% cut-off, there are three false negatives
and there is no correct identification of positive animals using this assay. Since the
elephant serum may react differently with the monoclonal, an ROC curve analysis of the
elephant samples was performed. This resulted in a sensitivity of 100% with a
95% confidence interval of (30.5 to 100), but a specificity of only 20% with a
95% confidence interval of (5.9 to 43.7). The best cut-off that could be determined was
at 0% inhibition which essentially renders the CI ELISA inappropriate for this species
under these assay conditions.
Table 3-3. Disagreement between PRNT results and CI ELISA results of the elephant
Elephant sample # Date sample taken WNV PRNT titer % Inhibition of MAb Agreement
1 1/30/2003 <10 0 Yes
2 1/30/2003 <10 0 Yes
3 1/30/2003 <10 0 Yes
4 1/30/2003 <10 0 Yes
5 1/30/2003 <10 0 Yes
6 1/30/2003 1:10 0 No
7 1/30/2003 <10 22.7 Yes
8 4/8/2003 <10 0 Yes
9 4/8/2003 <10 0 Yes
10 4/8/2003 <10 0 Yes
11 4/8/2003 >40 0 No
12 4/8/2003 <10 0 Yes
13 4/8/2003 <10 0 Yes
14 4/8/2003 <10 0 Yes
1 4/11/2003 <10 0 Yes
2 4/11/2003 <10 6.8 Yes
4 4/11/2003 <10 0 Yes
6 4/11/2003 <10 0 Yes
15 4/11/2003 >40 9.1 No
7 4/11/2003 <10 11.5 Yes
4 10/16/2003 <10 0 Yes
4 10/30/2003 <10 0 Yes
4 11/3/2003 <10 0 Yes
Historically, there had been no arboviruses described in any aquatic species. There
have been recent reports of alphaviruses in rainbow trout (Oncorhynchus mykiss),25
Atlantic salmon (Salmo salar),27 and southern elephant seals (Mirounga leonina).19 This
latter finding is of particular interest since the virus was isolated from the elephant seal
louse (Lepidophthirus macrorhini) which routinely infests the population of southern
elephant seals. It has been suggested that this new virus is transmitted by the louse,
primarily since the island has no mosquitoes and the seals are only rarely affected with
ticks. There was a report of a captive killer whale clinically affected with SLEV which
later died of the disease.8 Confirmation was made using virus isolation. More recently,
several captive harbor seals (Phoca vitulina) and a Hawaiian monk seal (Monachus
schauinslandi) were clinically affected with WNV.11,12,23 These seals exhibited clinical
signs including weakness, head tremor, anorexia, and abnormal swimming. Lesions
included a nonsuppurative encephalitis, meningitis, and myocarditis.
West Nile virus has been reported in many other wildlife and domestic species.
Clinically affected terrestrial mammalian species include squirrels, bats, chipmunks,
rabbits, raccoons, dogs, and cats.2 '413'15'20 There have been reports of several canids with
clinical WNV infection.18 A wolf pup and an adult 8-year old dog exhibited clinical
signs including anorexia, weakness, ataxia, ptyalism, head tilt, and blindness.18 The dog
had a concurrent immune-mediated thrombocytopenia which may have contributed to
disease susceptibility. Many more mammalian species have seroconverted to WNV
without development of clinical disease.
Reptiles and amphibians are also susceptible to WNV. Experimentally infected
lake frogs (Rana ridibunda) not only developed a high titer, but were able to transmit the
virus to Culexpipiens.16 More recently alligators have exhibited clinical signs with
WNV infection.22 Clinical signs seen in these animals included "star gazing", ataxia, and
Our CI ELISA screening of blinded samples did not identify any positive marine
mammals. Early experience with WNV suggests that pinnipeds may be the marine
mammal group most affected. Our sampling primarily consisted of cetacean and sirenian
sampling with few pinniped samples. Although it would appear that cetaceans are most
likely not affected by WNV, the SLEV infection reported previously in a captive killer
whale illustrates that such infections cannot be entirely ruled out.8
Several WNV and SLEV positive elephants were identified during the screenings
performed in this current study. All of these elephants have been in the U.S. for several
years, removing the likelihood of exposure prior to arrival into the U.S. There are no
published reports identifying seropositive elephants. There were no clinical signs
reported in any of the elephants with elevated titers. It would appear, based on this
preliminary data, that WNV infections remain unapparent in Asian elephants.
Previous CI ELISA formats described in the literature relied on the use of only one
monoclonal with sensitivities and specificities of 87% and 86%, respectively.6'7 The
competitive inhibition format with three monoclonals was investigated to enhance the
performance of this test. The surface located WNV E-protein has multiple functions
associated with virulence which include cell attachment and invasion. Complete
neutralization is difficult to achieve with only one monoclonal. Hence it was
hypothesized that multiple monoclonals that neutralized or reacted with this large protein
would be more correlated with an ELISA format. A competitive format using Japanese
encephalitis virus demonstrated 82.1% and 100% sensitivity and specificity when
compared to a neutralization format.10 In studies done by Beasley et al., three mouse
monoclonal antibodies were tested for their reactivity with the WNV E-protein domain
III.5 Results indicated that all three monoclonal antibodies strongly neutralized the virus.
However, no enhancement was observed regarding inhibition in the ELISA format. The
monoclonal 7H2 demonstrated reaction at or near the level of the monoclonal
combination and the benefit of using all three was not evident. The protocol was changed
to use monoclonal 7H2 only for all the future CI assays that were performed.
The original CI assay used mouse-brain antigen which resulted in problematic
results. There was a high degree of non-specific binding that was interfering with the
results. In addition, the inter-assay variation was quite high. Other antigens were tested
(data not shown) but were cost prohibitive. The WNV chimera antigen was far superior
to the mouse-brain antigen for several reasons. There was lower non-specific binding,
and had better reactivity antigen specific sera and monoclonals than the mouse-brain
Two species were used to determine a cut-off for the CI ELISA. A domestic
species, the horse, was selected because they are clinically affected with disease.
Alligators were chosen because they represented a wild species which also develops
clinical disease. In addition, there were many representative samples from these two
There was a good degree of consistency using this assay based on the CV results.
The cut-off value selected using the ROC curve analysis allowed identification of the
area on the curve with the highest sensitivity and specificity. By this analysis, this assay
has higher specificity than sensitivity. Using the ROC, this value can be increased by
changing the cut-off value. The specificity of the test would be decreased. However,
since the PRNT is the gold-standard, confirmatory testing would then be used to validate
the CI ELISA results in a diagnostic laboratory setting.
When a cut-off value using the percent inhibition is calculated, the sensitivity
remains nearly identical to that obtained using the OD cut-off. The specificity, however,
improves to 100%. Using the percent inhibition cut-off is more accurate, not only based
on these results, but also allows for the inter-assay variation that occurs between plates.
Therefore, it is recommended that calculation of percent inhibition be based on the
negative control of each plate.
Different species performed very differently on the assay and the assay should be
validated for each species. Although the original intention was for the CI ELISA to be
used as a screening assay, it is a test that has a high degree of specificity rather than a
high degree of sensitivity. It is clear, based on these findings, that the assay would need
to be validated for each individual species or group of animals prior to use as a screening
test. The assay performed moderately well on horse samples. Most horses today are
vaccinated against WNV and it remains unclear how this may be affecting the results.
The assay performed extremely well for alligator samples, giving 100% specificity and
sensitivity. All the positive alligators were clinically ill and likely had high viremic
levels. Based on this preliminary data, this may be an excellent assay for alligators and
demonstrates that both serum and plasma can be used. Further sampling should be
performed using wild-exposed alligators to determine whether the test will be sensitive
enough to identify those animals.
Cats are another group of animals where WNV may be of clinical importance and
serosurvey value. It has been demonstrated experimentally that domestic cats develop
higher viremic levels than most species, ranging from 103.2 to 104.0 PFU/mL.4 In
addition, they have been shown to be clinically affected by WNV and have the ability to
become infected through the ingestion of infected mice. The viremia in the cats is not
consistently above the 105 PFU/ml threshold necessary for the host to transmit the virus
to mosquitoes.9 However, this level is only an average and can vary between mosquito
species, therefore, cats could play a minor role as an amplifying host. The exotic cat
sampling consisted of only 5 samples. This is not enough to determine the validity of the
test. The preliminary results show 100% agreement with the PRNT and more samples
are necessary to further validate this format as a potential screening tool for this family.
The domestic cat samples obtained were all PRNT negative, therefore, no conclusions
could be made concerning the assay. A separate study concerning cats and WNV is
currently in progress.
There were not enough non-manatee marine mammal sample numbers to determine
whether or not this test will be of value in those species. And of those, there was only
one PRNT positive animal. In general, marine mammals do not seem to have significant
exposure to WNV. The only group of marine mammals where WNV has been a reported
problem, and could continue to be in the future, has been the captive pinnipeds.11,12'23 If
more of these infections become apparent, these samples could be used to validate the
test for pinnipeds.
As was the case with the cats, all of the manatees tested on the PRNT assay were
negative. Therefore, this assay is not currently validated for this species given that no
animals confirmed with WNV exposure were found in our PRNT confirmed dataset.
Most manatees did stay under the -28% cut-off that was developed using the horse and
alligator samples. If used as a screening assay, those few animals that tested above the
cut-off level would have needed to be confirmed using the PRNT assay, where they
would have been found negative.
There was no discernable method to develop a cut-off value with the elephant
samples. There were numerous false positives and false negatives. It is unclear why the
elephants reacted so poorly with this assay compared to other species. In order to verify
whether the PRNT assay results are accurate, rescreening of the elephants is necessary.
Further validation with a Western blot format may be necessary. If they are confirmed,
then this assay should not be considered for screening elephants and only serves to
illustrate the importance of validating the ELISA for each individual group of animals.
USE OF A COMPETITIVE INHIBITION ENZYME LINKED IMMUNOSORBENT
ASSAY FOR DETECTING WEST NILE VIRUS EXPOSURE IN A POPULATION OF
West Nile virus (WNV) is an arthropod-borne virus introduced to the New World
in 1999 that causes neurological disease primarily in birds, horses, and humans.1 Since
its initial appearance, this virus has spread rapidly across the North American continent.2
Thus far, harbor seals, monk seals, alligators, gray squirrels, alpacas, llamas, a wolf, and
one dog have all been clinically affected by natural infections.35'7-10'205 A wide range of
other animals has seroconverted to WNV with unknown clinical status. During
2003, WNV was established in much of the U.S., affecting over 9000 people with WNV
encephalitis and WNV fever. The actual infection and disease status has not been
completely assessed for many species. West Nile virus has been identified several
species of bats including big brown bats (Eptesicusfuscus) and little brown bats (Myotis
lucifugus).207 Serosurveys have been performed on some species of bats (Eptesicus
fuscus, M. lucifugus, and M. septentriotalis) indicating a low level of seropositivity
(2.4%) to WNV.270 No studies have been performed with the species in this study or in
bats in Florida. A competitive-inhibition enzyme-linked immunosorbent assay
(CI ELISA) has been developed (see chapter 3) to detect seroconversion to WNV for
various wildlife and domestic species. Our specific aim is to perform a serosurvey of
several species of captive bats and determine their exposure levels to WNV.
Materials and Methods
Bat plasma samples were obtained from routine blood samples taken from bats at
the Lubee bat conservancy (Gainesville, FL). Although these animals are captive, they
are kept outside in pens, and consequently are exposed to mosquitoes. A total of
39 animals representing five species were sampled (Table 4-1).
Table 4-1. Sample breakdown of the bat species tested
Species Number sampled
Pteropus vampyrus (Large flying fox) 18
Pteropuspumilus (Little golden-mantled flying fox) 6
Pteropus hypomelanus (Island flying fox) 5
Pteropus giganteus (Indian flying fox) 5
Pteropus poliocephalus (Gray-headed flying fox) 5
Samples were tested using the CI-ELISA as previously described. Briefly, a
96-well Immulon 2HB plate (Fisher, Pittsburgh, PA) was coated with WNV recombinant
antigen and allowed to incubate overnight at 40C. The wells were blocked with 150 ptL
of 5% Blotto for one hour at room temperature. The plasma samples were diluted
1:2 and added to the plate with monoclonal antibody 7H2 (Bioreliance, Rockville, MD).
The plate was incubated for one hour at 370C. Following incubation, a horseradish
peroxidase-conjugated goat anti-mouse IgG antibody (Amersham Biosciences,
Piscataway, NJ) was added. The plate was incubated for 30 minutes at 370C after which
3, 3', 5, 5'-tetramethylbenzidine solution (TMB) (KPL, Gaithersburg, MD) was added to
each well and the plate allowed to develop for 15 minutes at room temperature in a dark
drawer. The reaction was stopped after 15 minutes with 1% HC1 solution and the plate
read at 450 nm using a Bio-Tek plate reader (Bio-Tek Instruments, Winooski, VT).
Positive samples were determined by measuring the percent inhibition of the negative
control. The formula used for this was [100-(sample OD/negative control OD*100)].
Greater than 28% inhibition was considered positive.
Results showed that a majority of animals tested above the cut-off OD (Table 4-2).
The average OD of all samples was 0.396 with an average percent inhibition of 55%.
Even assuming a cut-off value of 50% or higher, this would still result in 32/39 with a
positive interpretation. Although P. vampyrus has a lower prevalence, this may be due to
the higher number of samples tested for that species.
Table 4-2. Competitive inhibition ELISA results on bats tested showing a large
percentage of positive samples
Species Number testing Percent positive
P. vampyrus 13/18 72.2
P. pumilus 6/6 100
P. hypomelanus 5/5 100
P. giganteus 5/5 100
P. poliocephalus 5/5 100
TOTAL 34/39 87.17
Although many of these bats tested well above the cut-off, this is the highest degree
of seropositivity in any of the species tested thus far. Either these are true positive
samples and there is a high degree of subclinical exposure in this species in Florida or
there is a very high degree of background reactivity in bat testing. None of the bats are
reported to have shown any clinical signs. As no overt clinical signs were reported in any
of the animals, it is possible that most infections are unapparent or this test is
inappropriate for use in this species.
Based on testing of this assay in multiple species, identification of true positive and
negative based on the gold standard, PRNT will be required to determine the true
prevalence of WNV in bats in Florida. Validation of this CI ELISA using bat plasma
should be performed to optimize the test for bat samples. Further testing of these samples
will be performed using virus neutralization to confirm infection of bats in Florida with
The purpose of this study was to develop a rapid screening assay to test numerous
wildlife and domestic species for WNV exposure. This assay could then be used to
screen a broad range of species for WNV exposure. Rapid, broad-based screening assays
require a test to be highly sensitive and easy to use. The PRNT assay is time consuming
(a minimum of 4 days) and requires the use of live virus and a BSL-3 laboratory in which
to perform it.a This CI ELISA has the benefit of taking only two days to perform and
does not require the use of a BSL-3 laboratory.
The mouse brain WNV antigen, a commercially available inactivated antigen,
while having acceptable antigen specific reactivity, proved unreliable. The antigen
resulted in high interassay variation and high non-specific binding (NSB). Because of the
requirement for live virus and live animal technology, this antigen is expensive and
difficult to obtain. The WNV chimera antigen was selected for use in this assay because
it was far superior to the mouse brain antigen. There was lower NSB and it had improved
reactivity with antigen-specific sera and monoclonals than the mouse brain antigen.
The two species used to determine a cut-off between negative and positive for the
CI ELISA were the horse and the alligator. Both of these species are clinically affected
by the disease and represent domestic and wild species. Two methods were tested to
establish a cut-off value; an OD cut-off, and a percent inhibition cut-off. The percent
a Centers for Disease Control and Prevention and National Institutes of Health. Biosafety in microbiological
and biomedical laboratory (BMBL). 4th edition. Washington: U.S. Government Printing Office; 1999.
inhibition cut-off value provided a better degree of sensitivity and controlled for
Using the ROC analysis, the cut-off value can be modified to allow for improved
sensitivity. If an assay has increased sensitivity, specificity will decrease and more false
positives will occur which will need to be confirmed using the PRNT assay. However,
since the PRNT is the gold-standard, confirmatory testing would then be used to validate
the CI ELISA results in a diagnostic laboratory setting, irrespective of CI ELISA
Of the various species tested, only the horses and the alligators provided enough
positive and negative samples for the assay to be validated. The assay resulted in
excellent sensitivity and specificity when tested with the alligator samples. All the
positive alligators were clinically ill and likely had high viremic levels. Based on this
preliminary data, this is an excellent assay for the alligator population and demonstrates
that both serum and plasma can be used. Further sampling should be performed using
wild-exposed alligators to determine whether the test will be sensitive enough to identify
those animals. The horses samples, when tested on the assay, had better specificity than
sensitivity. Most horses today are vaccinated against WNV and it remains unclear how
this may affect overall test performance. Irrespective this assay needs validation for each
individual species for use.
The marine mammal CI ELISA screening did not result in any positive marine
mammals. The only positive sample tested on this ELISA was a clinically ill harbor seal
sample. Because of the lack of positive samples, validation was not possible for this
group of animals. The only group of marine mammals where WNV has been a reported
problem are the pinnipeds.3'5'6 If more of these infections become apparent, these
samples could be used to validate the test for the pinnipeds. All of the manatees tested on
the PRNT assay were negative. Most manatees did stay under the -28% cut-off that was
developed using the horse and alligator samples. If used as a screening assay, any
animals testing above this cut-off require confirmation with the PRNT assay.
The elephant screening resulted in several positive elephants for both WNV and
SLEV exposure. One elephant was positive for both. None of these elephants were
clinically ill; therefore, it appears that WNV infection may not be a concern for this
species. Using an ROC curve, it was apparent that this version of the CI ELISA is does
not agree well with the PRNT testing in this species.
The cat screening included only a few positive exotic cat samples, but preliminary
results demonstrate 100% agreement with the PRNT. More samples are necessary to
validate this format as a screening tool for this exotics. A separate study concerning cats
and WNV is currently in progress.
The bats screened had a large number of potentially positive animals in their
population. These animals are very commonly exposed to mosquitoes and their exposure
may be similar to that of birds. However, this assay requires validation for each species,
so further testing of these samples is required before we can confirm whether these bats
have been exposed to WNV.
Few animals were SLEV positive. Therefore the extent, if any, of cross-reactivity
could not be determined. The degree of specificity may change when testing multiple
species that have a high degree of positivity for SLEV.
Although the original intention was for the CI ELISA to be used as a screening
assay, this project indicates that the test has a high degree of specificity rather than a high
degree of sensitivity. The ROC curve can be used to adjust the cut-off value and increase
sensitivity if desired. This assay reacts differently across species and requires validation
on each individual species before any broad-based testing applications can be done.
TESTING OF VERO CELL ANTIGEN ON THE CI ELISA
Inactivated West Nile virus (WNV) Vero cell antigen (gift from Dr. Elliot
Jacobsen) was tested for reactivity on the CI ELISA. As with the mouse brain WNV
antigen, a checkerboard plate was designed using Vero cell antigen dilutions of
1:250, 1:500, 1:1000, and 1:2000. Based on previous results, only the 7H2 monoclonal
was used at a dilution of 1:4000. The remainder of the CI ELISA plate was run as
Results indicated that the Vero cell antigen outperformed any results previously
obtained from the mouse brain antigen. In addition, the plate development occurred
much faster than expected and the reaction needed to be stopped at 10 minutes as
compared to 15 minutes with the mouse brain antigen. The Vero cell antigen was cost
prohibitive and as a result, we did not continue to pursue this antigen.
Vero at 1:250 Vero at 1:500 Vero at Vero at Mouse Brain Mouse Brain
1:1000 1:2000 at 1:50 at 1:100
Figure A-1. Composite graph comparing the results obtained using the Vero cell antigen
and the mouse brain antigen. Positive, negative, and NSB controls were
tested on all plates.
- _- Positive
OPTIMIZATION OF THE CI ELISA USING THE CHIMERIVAX ANTIGEN
A checkerboard plate was designed to determine the optimal dilution for the new
ChimeriVax antigen. Dilutions were tested at 1:1, 1:2, 1:8, 1:16, 1:32, 1:50, 1:100,
1:200, 1:400, 1:800, and 1:1600. Positive, negative, 7H2, blank, and NSB controls were
tested for all dilutions.
1 A A
N' N i N ti ge dl t on
N. N. N* NK* .N
ChimeriVax antigen dilutions
Figure B-1. Competitive inhibition ELISA plate coated with the ChimeriVax antigen.
Positive, negative, 7H2, and NSB controls were tested.
The assay performed well (Figure B-l) using the ChimeriVax antigen until the
1:32 dilution, where levels began to drop off. Based on these results, a dilution of
1:16 was selected for all future plates using this antigen.
As with the mouse brain WNV antigen, all three monoclonal antibodies were tested
using the antibodies alone at a 1:1000 dilution (3A3 and 5H10) and a 1:4000 dilution
(7H2). A combination of all three antibodies together (MAb) was also tested. Positive,
negative, blank, and NSB controls were run on all plates.
As with the mouse brain WNV antigen, the optical densities of the 7H2 and the
MAb were closely associated (Figure B-2). In contrast, the 3A3 and the 5H10 resulted in
lower optical densities. As a result, the 1:4000 dilution of the 7H2 was continued for use
in all future assays using the ChimeriVax antigen.
0.4 --- Negative
7H2 3A3 5H10 3Mabs
Figure B-2. Competitive inhibition ELISA plate coated with the ChimeriVax antigen at
a dilution of 1:16. Monoclonals 7H2 (1:4000), 3A3 (1:1000), and 5H10
(1:1000) were tested against this antigen for reactivity.
Comparison of Conjugate Diluents
The conjugate diluent originally used for the mouse brain antigen was 5% Blotto
with normal goat serum. As noted in Chapter 3, the original mouse brain antigen resulted
in very high non-specific binding (NSB). Because of this, several blocking agents were
used throughout the assay to control the NSB. Using the Blotto was an attempt to reduce
the NSB. Once the antigen was changed, testing was performed to determine whether
this extra blocking step should be continued or should be eliminated.
A plate was coated with the ChimeriVax antigen at a 1:16 dilution. Standard
controls were tested including positive, negative, NSB, and blanks. The left half of the
plate was tested using PBST only as the diluent for the conjugated antibody. The right
half of the plate used the conjugate diluent as previously described. In addition, the
samples and controls were run at dilutions ranging from 1:2 to 1:16 to determine optimal
Using conjugated antibody diluted in PBST alone resulted in improved results with
no increase in NSB. The NSB average for the Blotto plate was .0485 while the NSB for
the plate using the PBST only was .0455. For the positive control, there was more
reactivity when using the non-blocked conjugate (Figure B-3). In addition, the dilution
of 1:2 resulted in the best reactivity. Results were similar for the negative control (Figure
Using these results, the procedure for the CI ELISA was modified to dilute the
conjugated antibody in PBST only instead of the blocking solution. All CI ELISA
procedures in the body of the thesis were performed using PBST diluent alone.
NVSL + Control
at 1:2 at 1:4 at 1:8
Figure B-3. Testing the NVSL positive control at varying dilutions to compare diluting
the conjugated antibody in a blocking solution versus using PBST alone.
at 1:2 at 1:4 at 1:8
Figure B-4. Testing the NVSL negative control at varying dilutions to compare diluting
the conjugated antibody in a blocking solution versus using PBST alone.
Although the ChimeriVax antigen was performing well, the negative control was
not optimal and several additional negative controls were tested. Serum from several
screened negative horses was collected and pooled to serve as a negative control. This
control was tested against the positive, negative, NSB, and MAb controls and against two
known positive animals; India, a borderline PRNT positive elephant at 1:10 and Frankie a
PRNT positive horse.
India Frankie NVSL IgG negative
Figure B-5. Comparison of known positive serum samples and their percent inhibition
using the new negative control and the new recombinant antigen.
The pooled negative control (Figure B-5) clearly identified Frankie as a positive
horse with over 70% inhibition and placed India at the borderline. As a comparison, the
NVSL IgG negative control that had been used has quite a high inhibition compared to
the new control. This negative control was adopted for use in all future CI ELISA
Second ChimeriVax Antigen Batch
Once the CI ELISA protocol had been finalized, we requested a new shipment of
antigen. This new shipment proved to be a different dilution that the previous one.
Therefore, a new checkerboard plate was run to determine the level of optimal
performance. This procedure is described in Chapter 3 and was determined to be 1:64.
Therefore, all validation procedures performed in the thesis were done with the new batch
of antigen at a 1:64 dilution.
LIST OF REFERENCES
1. MMWR. Outbreak of West Nile-like viral encephalitis-New York, 1999.
MMWR Morb Mortal Wkly Rep. 1999;48:845-849.
2. MMWR. Provisional surveillance summary of the West Nile virus
epidemic-United States, January-November 2002. MMWR Morb Mortal Wkly
3. Gentz EJ, Richard MJ. Infection in two harbor seals (Phoca vitulina) with West
Nile virus [abstract]. In: Proceedings of the International Association for Aquatic
AnimalMedicine, Galveston, TX; 2004:160-161.
4. Kutzler MA, Bildfell RJ, Gardner-Graff KK, Baker RJ, Delay JP, Mattson DE.
West Nile virus infection in two alpacas. JAm Vet MedAssoc. 2004;225:921-924.
5. Stremme DW, Duncan A. Clinical signs of West Nile flavivirus
polioencephalomyelitis in a harbor seal (Phoca vitulina) [abstract]. In:
Proceedings of the International Association for Aquatic Animal Medicine,
Waikoloa, HI; 2003:34-38.
6. Dalton LM, Dickerson S, Wigdahl D. A serosurvey for West Nile virus at
SeaWorld San Antonio [abstract]. In: Proceedings of the International
Association for Aquatic Animal Medicine, Galveston, TX; 2004: 148-150.
7. Miller DL, Mauel MJ, Baldwin C, Burtle G, Ingram D, Hines ME, Frazier KS.
West Nile virus in farmed alligators. Emerg Infect Dis. 2003;9:794-799.
8. Heinz-Taheny KM, Andrews JJ, Kinsel MJ, Pessier AP, Pinkerton ME,
Lemberger KY, Novak RJ, Dizikes GJ, Edwards E, Komar N. West Nile virus
infection in free-ranging squirrels in Illinois. J Vet Diagn Invest. 2004;16:186-
9. Lichtensteiger CA, Heinz-Taheny K, Osborne TS, Novak RJ, Lewis BA, Firth
ML. West Nile virus encephalitis and myocarditis in wolf and dog. Emerg Infect
10. Yaeger M, Yoon KJ, Schwartz K, Berkland L. West Nile virus
meningoencephalitis in a Suri alpaca and Suffolk ewe. J Vet Diagn Invest.
11. Weston JH, Welsh MD, McLoughlin MF, Todd D. Salmon pancreas disease
virus, an alphavirus infecting farmed Atlantic salmon, Salmo salar L. Virology.
12. Villoing S, Bearzotti M, Chilmonczyk S, Castric J, Bremont M. Rainbow trout
sleeping disease virus is an atypical alphavirus. J Virol. 2000;74:173-183.
13. Ward JR, Lafferty KD. The elusive baseline of marine disease: Are diseases in
ocean ecosystems increasing? PLoSBiol. 2004;2:e120.
14. Linn ML, Gardner J, Warrilow D, Darnell GA, McMahon CR, Field I, Hyatt AD,
Slade RW, Suhrbier A. Arbovirus of marine mammals: a new alphavirus isolated
from the elephant seal louse, Lepidophthirus macrorhini. J Virol. 2001;75:4103-
15. Mccaig LF, Janowski HT, Gunn RA, Tsai TF. Epidemiologic aspects of a St.
Louis encephalitis outbreak in Fort Walton Beach, Florida in 1980. Am J Trop
16. Martin DA, Biggerstaff BJ, Allen B, Johnson AJ, Lanciotti RS, Roehrig JT. Use
of immunoglobulin M cross-reactions in differential diagnosis of human flaviviral
encephalitis infections in the United States. Clin Diagn Lab Immunol.
17. Cornaglia E, Rebora L, Gili C, Di Guardo G. Histopathological and
immunohistochemical studies on cetaceans found stranded on the coast of Italy
between 1990 and 1997. J VetMedA-PhysiolPathol Clin Med. 2000;47:129-142.
18. Jauniaux T, Petitjean D, Brenez C, Borrens M, Brosens L, Haelters J, Tavemier T,
Coignoul F. Post-mortem findings and causes of death of harbour porpoises
(Phocoenaphocoena) stranded from 1990 to 2000 along the coastlines of
Belgium and northern France. JComp Pathol. 2002;126:243-253.
19. Ross HM, Foster G, Reid RJ, Jahans KL, MacMillan AP. Brucella species
infection in sea-mammals. Vet Rec. 1994;134:359.
20. Ewalt DR, Payeur JB, Martin BM, Cummins DR, Miller WG. Characteristics of a
Brucella species from a bottlenose dolphin (Tursiops truncatus). J Vet Diagn
21. Cloeckaert A, Grayon M, Grepinet O, Boumedine KS. Classification of Brucella
strains isolated from marine mammals by infrequent restriction site-PCR and
development of specific PCR identification tests. Microbes Infect. 2003;5:593-
22. Van Bressem MF, Van Waerebeek K, Raga JA, Godfroid J, Brew SD, MacMillan
AP. Serological evidence of Brucella species infection in odontocetes from the
south Pacific and the Mediterranean. Vet Rec. 2001;148:657-661.
23. Nielsen O, Stewart REA, Nielsen K, Measures L, Duignan P. Serologic survey of
Brucella spp. antibodies in some marine mammals of North America. J WildlDis.
24. Ohishi K, Zenitani R, Bando T, Goto Y, Uchida K, Maruyama T, Yamamoto S,
Miyazaki N, Fujise Y. Pathological and serological evidence of Brucella-infection
in baleen whales (Mysticeti) in the western North Pacific. Comp Immunol
Microbiol Infect Dis. 2003;26:125-136.
25. Ross HM, Jahans KL, MacMillan AP, Reid RJ, Thompson PM, Foster G.
Brucella species infection in North Sea seal and cetacean populations. Vet Rec.
26. Maratea J, Ewalt DR, Frasca Jr S, Dunn JL, DeGuise S, Szudlarek L, St.Aubin
DJ, French RA. Evidence of Brucella sp. infection in marine mammals stranded
along the coast of southern New England. JZoo WildlMed. 2003;34:256-261.
27. Foster G, MacMillan AP, Godfroid J, Howie F, Ross HM, Cloeckaert A, Reid RJ,
Brew S, Patterson IAP. A review of Brucella sp infection of sea mammals with
particular emphasis on isolates from Scotland. VetMicrobiol. 2002;90:563-580.
28. Foster G, Jahans KL, Reid RJ, Ross HM. Isolation of Brucella species from
cetaceans, seals and an otter. VetRec. 1996;138:583-586.
29. Tryland M, Derocher AE, Wiig O, Godfroid J. Brucella sp antibodies in polar
bears from Svalbard and the Barents Sea. J WildlDis. 2001;37:523-531.
30. Miller WG, Adams LG, Ficht TA, Cheville NF, Payeur JP, Harley DR, House C,
Ridgway SH. Brucella-induced abortions and infection in bottlenose dolphins
(Tursiops truncatus). JZoo Wildl Med. 1999;30:100-110.
31. Gonzalez L, Patterson IA, Reid RJ, Foster G, Barberan M, Blasco JM, Kennedy
S, Howie FE, Godroid J, MacMillan AP, Schock A, Buxton D. Chronic
meningoencephalitis associated with Brucella sp infection in live-stranded striped
dolphins (Stenella coeruleoalba). J Comp Pathol. 2002;126:147-152.
32. Sohn AH, Probert WS, Glaser CA, Gupta N, Bollen AW, Wong JD, Grace EM,
McDonald WC. Human neurobrucellosis with intracerebral granuloma caused by
a marine mammal Brucella spp. EmergInfect Dis. 2003;9:485-488.
33. Brew SD, Perrett LL, Stack JA, MacMillan AP, Staunton NJ. Human exposure to
Brucella recovered from a sea mammal. Vet Rec. 1999;144:483.
34. Kennedy S. Morbillivirus infections in aquatic mammals. J Comp Pathol.
35. Di Guardo G, Marruchella G, Agrimi U, Kennedy S. Morbillivirus infections in
aquatic mammals: A brief overview. J Vet MedA Physiol Pathol Clin Med.
36. Duignan PJ, House C, Odell DK, Wells RS, Hansen LJ, Walsh MT, StAubin DK,
Rima BK, Geraci JR. Morbillivirus infection in bottlenose dolphins: Evidence for
recurrent epizootics in the western Atlantic and Gulf of Mexico. Mar Mamm Sci.
37. Geraci JR. Clinical investigation of the 1987-1988 mass mortality of bottlenose
dolphins along the U.S. central and w,,/h Atlantic coast. Washington DC. Final
report to the National Marine Fisheries Service and U.S. Navy Office of Naval
Research and Marine Mammal Commission; 1989.
38. Lipscomb TP, Schulman FY, Moffett D, Kennedy S. Morbilliviral disease in
Atlantic bottlenose dolphins (Tursiops truncatus) from the 1987-1988 epizootic. J
39. Grachev MA, Kumarev VP, Mamaev LV, Zorin VL, Baranova LV, Denikina NN,
Belikov SI, Petrov EA, Kolesnik VS, Kolesnik RS, Dorofeev VM, Beim AM,
Kudelin VN, Nagieva FG, Sidorov VN. Distemper virus in Baikal seals. Nature.
40. Follmann EH, Garner GW, Evermann JF, McKeirnan AJ. Serological evidence of
morbillivirus infection in polar bears (Ursus maritimus) from Alaska and Russia.
41. Garner GW, Evermann JF, Saliki JT, Follmann EH, McKeirnan AJ. Morbillivirus
ecology in polar bears (Ursus maritimus). Polar Biol. 2000;23:474-478.
42. Osterhaus AD, Vedder EJ. Identification of virus causing recent seal deaths.
43. Osterhaus ADME, Deswart RL, Vos HW, Ross PS, Kenter MJH, Barrett T.
Morbillivirus infections of aquatic mammals: Newly identified members of the
genus. Vet Microbiol. 1995;44:219-227.
44. Domingo M, Vilafranca M, Visa J, Prats N, Trudgett A, Visser I. Evidence for
chronic morbillivirus infection in the Mediterranean striped dolphin (Stenella
coeruleoalba). Vet Microbiol. 1995;44:229-239.
45. Kennedy S, Kuiken T, Ross HM, Mcaliskey M, Moffett D, Mcniven CM,
Carole M. Morbillivirus infection in two common porpoises (Phocoena
phocoena) from the coasts of England and Scotland. Vet Rec. 1992; 131:286-290.
46. Taubenberger JK, Tsai M, Krafft AE, Lichy JH, Reid AH, Schulman FY,
Lipscomb TP. Two morbilliviruses implicated in bottlenose dolphin epizootics.
Emerg Infect Dis. 1996;2:213-216.
47. Mamaev LV, Denikina NN, Belikov SI, Volchkov VE, Visser IKG, Fleming M,
Kai C, Harder TC, Liess B, Osterhaus ADME, Barrett T. Characterisation of
morbilliviruses isolated from Lake Baikal seals (Phoca sibirica). Vet Microbiol.
48. Kennedy S, Kuiken T, Jepson PD, Deaville R, Forsyth M, Barrett T, van de Bildt
MW, Osterhaus AD, Eybatov T, Duck C, Kydyrmanov A, Mitrofanov I, Wilson
S. Mass die-off of Caspian seals caused by canine distemper virus. Emerg Infect
49. Krafft A, Lichy JH, Lipscomb TP, Klaunberg BA, Kennedy S, Taubenberger JK.
Postmortem diagnosis of morbillivirus infection in bottlenose dolphins (Tursiops
truncatus) in the Atlantic and Gulf of Mexico epizootics by polymerase chain
reaction-based assay. J WildlDis. 1995;31:410-415.
50. Osterhaus A, Groen J, Niesters H, van de BM, Martina B, Vedder L, Vos J, van
Egmond H, Abou-Sidi B, Barham ME. Morbillivirus in monk seal mass mortality.
51. Jensen T, van de BM, Dietz HH, Andersen TH, Hammer AS, Kuiken T,
Osterhaus A. Another phocine distemper outbreak in Europe. Science.
52. Muller G, Wohlsein P, Beineke A, Haas L, Greiser-Wilke I, Siebert U, Fonfara S,
Harder T, Stede M, Gruber AD, Baumgartner W. Phocine distemper in German
seals, 2002. Emerg Infect Dis. 2004; 10:723-725.
53. Cosby SL, McQuaid S, Duffy N, Lyons C, Rima BK, Allan GM, McCullough SJ,
Kennedy S, Smyth JA, McNeilly F, Craig C, Orvell C. Characterization of a seal
morbillivirus. Nature. 1988;336:115-116.
54. Kennedy S, Smyth JA, McCullough SJ, Allan GM, McNeilly F, McQuaid S.
Confirmation of cause of recent seal deaths. Nature. 1988;335:404.
55. Duignan P. Morbillivirus infections of marine mammals. In: Fowler ME, Miller
RE, eds. Zoo & WildAnimal Medicine.1999; Philadelphia: W.B. Saunders.
56. Saliki JT, Lehenbauer TW. Monoclonal antibody-based competitive enzyme-
linked immunosorbent assay for detection of morbillivirus antibody in marine
mammal sera. J ClinicalMicrobiol. 2001;39:1877-1881.
57. Saliki JT, Cooper EJ, Gustavson JP. Emerging morbillivirus infections of marine
mammals: development of two diagnostic approaches. Ann N YAcad Sci.
58. Kennedy-Stoskopf S. Viral diseases. In: Dierauf LA, Gulland FM, eds. CRC
Handbook ofMarine Mammal Medicine.2001; Boca Raton, FL: CRC Press LLC.
59. Odegaard OA, Krogsrud J. Rabies in Svalbard: Infection diagnosed in arctic fox,
reindeer and seal. VetRec. 1981;109:141-142.
60. Loewen K, Prins B, Philibert H. Rabies in a polar bear. Can VetJ. 1990;31:457.
61. Taylor M, Elkin B, Maier N, Bradley M. Observation of a polar bear with rabies.
62. Ray CG. Arthropod-bome and other zoonotic viruses. In: Ryan KJ, ed. .lher i
Medical Microbiology: An introduction to infectious diseases. 1994; Norwalk, CT:
Appleton and Lange.
63. Villoing S, Bearzotti M, Chilmonczyk S, Castric J, Bremont M. Rainbow trout
sleeping disease virus is an atypical alphavirus. J Virol. 2000;74:173-183.
64. Weston JH, Welsh MD, McLoughlin MF, Todd D. Salmon pancreas disease
virus, an alphavirus infecting farmed Atlantic salmon, Salmo salar L. Virology.
65. Buck C, Paulino GP, Medina DJ, Hsiung GD, Campbell TW, Walsh MT.
Isolation of St. Louis encephalitis virus from a killer whale. Clin Diagn Virol.
66. Inskeep W, Gardiner CH, Harris RK, Dubey JP, Goldston RT. Toxoplasmosis in
Atlantic bottlenose dolphins (Tursiops truncatus). J WildlDis. 1990;26:377-382.
67. Cruickshank JJ, Haines DM, Palmer NC, Staubin DJ. Cysts of a Toxoplasma-like
organism in an Atlantic bottlenose dolphin. Can VetJ. 1990;31:213-215.
68. Cole RA, Lindsay DS, Howe DK, Roderick CL, Dubey JP, Thomas NJ, Baeten
LA. Biological and molecular characterizations of Toxoplasma gondii strains
obtained from southern sea otters (Enhydra lutris nereis). JParasitol.
69. Hanni KD, Mazet JAK, Gulland FMD, Estes J, Staedler M, Murray MJ, Miller M,
Jessup DA. Clinical pathology and assessment of pathogen exposure in southern
and Alaskan sea otters. J WildlDis. 2003;39:837-850.
70. Migaki G, Allen JF, Casey HW. Toxoplasmosis in a California sea lion (Zalophus
californianus). Am Journal Vet Res. 1977;38:135-136.
71. Mikaelian I, Boisclair J, Dubey JP, Kennedy S, Martineau D. Toxoplasmosis in
beluga whales (Delphinapterus leucas) from the St Lawrence estuary: Two case
reports and a serological survey. J Comp Pathol. 2000; 122:73-76.
72. Miller MA, Sverlow K, Crosbie PR, Barr BC, Lowenstine LJ, Gulland FM,
Packham A, Conrad PA. Isolation and characterization of two parasitic protozoa
from a Pacific harbor seal (Phoca vitulina richardsi) with
meningoencephalomyelitis. JParasitol. 2001;87:816-822.
73. Buergelt CD, Bonde RK. Toxoplasmic meningoencephalitis in a West Indian
manatee. JAm VetMedAssoc. 1983;183:1294-1296.
74. Jardine JE, Dubey JP. Congenital toxoplasmosis in a Indo-Pacific bottlenose
dolphin (Tursiops aduncus). JParasitol. 2002;88:197-199.
75. Lindsay DS, Thomas NJ, Rosypal AC, Dubey JP. Dual Sarcocystis neurona and
Toxoplasma gondii infection in a Northern sea otter from Washington state, USA.
Vet Parasitol. 2001;97:319-327.
76. Gulland FMD, Haulena M, Dierauf LA. Seals and Sea lions. In: Dierauf LA,
Gulland FMD, eds. CRC Handbook ofMarine Mammal Medicine.2001; Boca
Raton, FL: CRC Press.
77. Dubey JP, Davis SW, Speer CA, Bowman DD, Delahunta A, Granstrom DE,
Topper MJ, Hamir AN, Cummings JF, Suter MM. Sarcocystis neurona sp
(Protozoa, Apicomplexa), the etiologic agent of equine protozoal
myeloencephalitis. JParasitol. 1991;77:212-218.
78. Lapointe JM, Duignan PJ, Marsh AE, Gulland FM, Barr BC, Naydan DK, King
DP, Farman CA, Huntingdon KAB, Lowenstine LJ. Meningoencephalitis due to a
Sarcocystis neurona-like protozoan in Pacific harbor seals (Phoca vitulina
richardsi). JParasitol. 1998;84:1184-1189.
79. Lindsay DS, Thomas NJ, Dubey JP. Biological characterisation of Sarcocystis
neurona isolated from a Southern sea otter (Enhydra lutris nereis). Int JParasitol.
80. Miller MA, Crosbie PR, Sverlow K, Hanni K, Barr BC, Kock N, Murray MJ,
Lowenstine LJ, Conrad PA. Isolation and characterization of Sarcocystis from
brain tissue of a free-living southern sea otter (Enhydra lutris nereis) with fatal
meningoencephalitis. Parasitol Res. 2001;87:252-257.
81. Rosonke BJ, Brown SR, Tornquist SJ, Snyder SP, Garner MM, Blythe LL.
Encephalomyelitis associated with a Sarcocystis neurona-like organism in a sea
otter. JAm VetMedAssoc. 1999;215:1839-1842.
82. Miller MA, Gardner IA, Kreuder C, Paradies DM, Worcester KR, Jessup DA,
Dodd E, Harris MD, Ames JA, Packham AE, Conrad PA. Coastal freshwater
runoff is a risk factor for Toxoplasma gondii infection of southern sea otters
(Enhydra lutris nereis). Int J Parasitol. 2002;32:997-1006.
83. Oshea TJ, Homer BL, Greiner EC, Layton AW. Nasitrema sp. associated
encephalitis in a striped dolphin (Stenella coeruleoalba) stranded in the Gulf of
Mexico. J WildlDis. 1991;27:706-709.
84. Parker GA, Migaki G, Walker WA. Cerebral trematodiasis in a dolphin. MilMed.
85. Dailey MD, Walker WA. Parasitism as a factor (?) in single standings of
southern California cetaceans. JParasitol. 1978;64:593-596.
86. Kvitek RG, DeGange AR, Beitler MK. Paralytic shellfish poisoning toxins
mediate feeding behavior of sea otter. Limnol Oceanogr. 1991;36:393-404.
87. Kvitek RG, Bowlby CE, Staedler M. Diet and foraging behavior of sea otters in
southeast Alaska. Mar Mamm Sci. 1993;9:168-181.
88. Geraci JR, Anderson DM, Timperi RJ, St.Aubin DJ, Early GA, Prescott JH, Mayo
CA. Humpback whales (Megaptera novaeangliae) fatally poisoned by
dinoflagellate toxin. Can JFish Aquat Sci. 1989;46:1895-1898.
89. Forcada J, Hammond PS, Aguilar A. Status of the Mediterranean monk seal
Monachus monachus in the Western Sahara and the implications of a mass
mortality event. Mar Ecol Prog Ser. 1999; 188:249-261.
90. O'Hara TM, O'Shea TJ. Toxicology. In: DieraufLA, Gulland FM, eds. CRC
Handbook ofMarine Mammal Medicine.2001; Boca Raton: CRC Press.
91. Bossart GD, Baden DG, Ewing RY, Roberts B, Wright SD. Brevetoxicosis in
manatees (Trichechus manatus latirostris) from the 1996 epizootic: gross,
histologic, and immunohistochemical features. Toxicol Pathol. 1998;26:276-282.
92. Oshea TJ, Rathbun GB, Bonde RK, Buergelt CD, Odell DK. An epizootic of
Florida manatees associated with a dinoflagellate bloom. Mar Mamm Sci.
93. Iverson F, Truelove J. Toxicology and seafood toxins: domoic acid. Nat Toxins.
94. Sobotka TJ, Brown R, Quander DY, Jackson R, Smith M, Long SA, Barton CN,
Rountree RL, Hall S, Eilers P, Johannessen JN, Scallet AC. Domoic acid:
Neurobehavioral and neurohistological effects of low-dose exposure in adult rats.
Neurotoxicol Teratol. 1996; 18:659-670.
95. Scholin CA, Gulland F, Doucette GJ, Benson S, Busman M, Chavez FP, Cordaro
J, DeLong R, De Vogelaere A, Harvey J, Haulena M, Lefebvre K, Lipscomb T,
Loscutoff S, Lowenstine LJ, Marin R, Miller PE, McLellan WA, Moeller PDR,
Powell CL, Rowles T, Silvagni P, Silver M, Spraker T, Trainer V, Van Dolah
FM. Mortality of sea lions along the central California coast linked to a toxic
diatom bloom. Nature. 2000;403:80-84.
96. Silvagni PA, Lowenstine LJ, Spraker T, Lipscomb TP, Gulland FM. Pathology of
domoic acid toxicity in California sea lions (Zalophus californianus). Vet Pathol.
97. Gulland FM, Haulena M, Fauquier D, Langlois G, Lander ME, Zabka T, Duerr R.
Domoic acid toxicity in Californian sea lions (Zalophus californianus): clinical
signs, treatment and survival. Vet Rec. 2002; 150:475-480.
98. Alexander JW, Solangi MA, Riegel LS. Vertebral osteomyelitis and suspected
diskospondylitis in an Atlantic bottlenose dolphin (Tursiops truncatus). J Wildl
99. Jacobs MS, McFarland WL, Morgane PJ. The anatomy of the brain of the
bottlenose dolphin (Tursiops truncatus). Rhinic lobe (Rhinencephalon): The
archicortex. Brain Res Bull. 1979;4 Suppl 1:1-108.
100. Marino L, Sudheimer K, McLellan WA, Johnson JI. Neuroanatomical structure of
the spinner dolphin (Stenella longirostris orientalis) brain from magnetic
resonance images. Anat Rec A Discov Mol Cell Evol Biol. 2004;279:601-610.
101. Marino L, Sudheimer KD, Pabst DA, McLellan WA, FilsoofD, Johnson JI.
Neuroanatomy of the common dolphin (Delphinus delphis) as revealed by
magnetic resonance imaging (MRI). Anat Rec. 2002;268:411-429.
102. Marino L, Murphy TL, Gozal L, Johnson JI. Magnetic resonance imaging and
three-dimensional reconstructions of the brain of a fetal common dolphin,
Delphinus delphis. Anat Embryol (Berl). 2001;203:393-402.
103. Taylor SM. The neurologic examination. In: Nelson RW, Couto CG, eds. Small
Animal Internal Medicine. 1998; St. Louis, MO: Mosby Inc.
104. Ridgway SH, Demski LS, Bullock TH, Schwanzel-Fukuda M. The terminal nerve
in odontocete cetaceans. Ann N YAcadSci. 1987;519:201-212.
105. Oelschlager HA. Early development of the olfactory and terminalis systems in
baleen whales. Brain Behav Evol. 1989;34:171-183.
106. Demski LS, Schwanzel-Fukuda M. The terminal nerve nervouss terminalis):
structure, function, and evolution. Introduction. Ann N YAcad Sci. 1987;519:ix-