Isolation of Saint Louis Encephalitis Virus from a Horse with Neurological Disease in Brazil

St. Louis encephalitis virus (SLEV) is a causative agent of encephalitis in humans in the Western hemisphere. SLEV is a positive-sense RNA virus that belongs to the Flavivirus genus, which includes West Nile encephalitis virus, Japanese encephalitis virus, Dengue virus and other medically important viruses. Recently, we isolated a SLEV strain from the brain of a horse with neurological signs in the countryside of Minas Gerais, Brazil. The SLEV isolation was confirmed by reverse-transcription RT-PCR and sequencing of the E protein gene. Virus identity was also confirmed by indirect immunofluorescence using commercial antibodies against SLEV. To characterize this newly isolated strain in vivo, serial passages in newborn mice were performed and led to hemorrhagic manifestations associated with recruitment of inflammatory cells into the central nervous system of newborns. In summary this is the first isolation of SLEV from a horse with neurological signs in Brazil.


Introduction
St. Louis encephalitis virus (SLEV) is a mosquito-borne virus that causes human and animal encephalitis in the Western hemisphere. SLEV is considered endemic in the Americas, with encephalitis cases being diagnosed from Canada to Argentina [1][2][3]. There is no vaccine or treatment available for St. Louis encephalitis.
SLEV is a single-stranded positive sense RNA virus, with approximately 50 nm in diameter and a genome of 11 kb. SLEV is a member of the Flavivirus genus in the Flaviviridae family, together with several important pathogens such as West Nile virus (WNV), Japanese encephalitis virus (JEV), Dengue virus (DENV), Yellow fever virus (YFV) and others [4,5]. Viral life cycle is enzootic and birds are the natural amplifying host [6]. Other vertebrates (e.g. wild animals, horses, and humans) are considered accidental/final hosts [7][8][9].
Human infections with SLEV are mostly asymptomatic. Infected individuals can present mild malaise or flu-like symptoms, especially young or middle-aged patients [6,10]. Severe cases are clinically characterized by high fever, neurological dysfunction, altered consciousness, and headache; which are accompanied by encephalitis or meningoencephalitis that affects more often the elderly [11][12][13]. Lethality rates in severe cases can reach 30%, and are associated to direct damage to the central nervous system (CNS) [3]. Acute illness can be followed by prolonged convales-cence with cognitive and psychosocial deficits for over a year [6,14]. Disease in wild or domestic animals has not been described, although many species are infected or are serologically positive for SLEV in endemic areas [6,[15][16][17][18][19].
SLEV has been detected in Brazil for over 40 years, isolated from arthropods [19] or by serological surveys in birds [20] and mammals [18,21]. SLEV was isolated from two patients in the Amazon region in 1970's [22,23] and isolated again from a dengue-suspected patient in Southeastern Brazil, in the early 2000's [2]. Interestingly, SLEV infections in humans were identified in southeast Brazil in the following years, under an outbreak of DENV-3, together with the first a human case of DENV-3 and SLEV co-infection [24,25].
Here we describe the first isolation of SLEV from a horse with neurological signs in Brazil. SLEV identity was confirmed by molecular and serological techniques, and by inoculation of newborn mice. Our findings highlight the importance of effective arboviral surveillance.

Ethics statement
Our animal study followed national guidelines (Law number 11.794, 8/10/2008), which governs the use of animals for experimental procedures. All experimental procedures were approved and complied with the University of Minas Gerais (UFMG) Committee for Ethics in Animal Experimentation (CETEA) regulations, under protocol number 163/2011.

Mice
Pregnant female mice were acquired from Centro de Bioterismo (CEBIO) of UFMG (Belo Horizonte, Brazil). Newborn Swiss mice (24 hours old) were used in animal model development experiments. All mice were kept under controlled temperature (23uC) with a strict 12 h light/dark cycle, food and water available ad libitum, under specific pathogen-free conditions, in the animal warren at the Departamento de Clínica e Cirurgia Veterinárias of UFMG.

Sample collection and PCR screening
Brain tissue from horses that presented neurological symptoms before death were sent to Laboratório de Saúde Animal at Instituto Mineiro de Agropecuária (LSA/IMA) in Belo Horizonte, Brazil. Samples that were PCR negative for Rabies virus were sent to Laboratório de Patologia Molecular at UFMG and stored at 280uC. Tissue samples were processed for RNA extraction and screened by nested RT-PCR. There were no tissue samples available for histopathology. Reaction parameters and primers were used as described by Ré and colleagues [26]. Primers used for the initial amplification were SLE 1497 (+) RRYATGGGYGAG-TATGGRACAG, SLE 2517 (2) CTCCTCCACAYTTYARTT-CACG, and primers for the final amplification were SLE (+) 2002 TGGAYTGGACRCCGGTTGGAAG and SLE (2) 2257 CCAATRGATCCRAARTCCCACG.

Sequencing and analysis
SLEV RT-PCR amplicon band was purified from an agarose gel and sequenced in Megabace 1000 sequencer. Edited sequences were aligned by CLUSTAL/W, using the BioEdit program, version 5.09. The resulting sequence was deposited in GenBank (accession number KF718857).

Virus isolation
A tissue fragment of the SLEV-positive brain was homogenized and clarified by centrifugation. The brain homogenate was inoculated on monolayers of the mosquito cell lineage C6/36 and cultures were monitored for cytopathic effect (CPE) daily. Viral stocks were collected for up to five days post-infection and reinoculated (500 mL of supernatant on a fresh culture) two times, for a total of three passages. The viral titer from supernatant of the third passage was determined by focus immunodetection assay (FIA) as previously described [32]. We obtained SLEV stocks at 1610 3 focus forming units, or FFU, per mL of culture supernatant.

Virus characterization in vivo
To characterize the isolated SLEV strain, 40 mL of SLEV stocks were inoculated in newborn mice by intracranial route (frontal left region of the brain). Homogenates of a pool of brain samples (10% in PBS) from the littermates were used for preparing the inoculum for the next passage. Infected newborn brains were collected at the onset of neurological disease or at day 7 post-infection, to produce new virus stocks or to process for histological analysis. Brain suspensions were passed seven times in newborn mouse brains and relevant experimental controls were maintained. Clinical alterations were assessed daily in newborns after each SLEV passage and representative pictures were taken.

Histology
Tissue samples, including brain, kidney, liver, lung, heart, and fragments of the thoracic and pelvic limbs, were obtained from two newborn mice at each SLEV passage, immediately fixed in 4% buffered formaldehyde, processed and embedded in paraffin. Tissue sections (4 mm thick) were stained with hematoxylin and eosin (HE), and examined under light microscopy. Micrographs were taken using a Spot Insight color camera coupled to Olympus BX41 microscope.
Immunofluorescence microscopy C6/36 cells were harvested and seeded in 24-well plates with gelatin-coated coverslips, and incubated at 28uC for at least 3 hours. Cells were infected with a virus stock derived from the 7 th SLEV passage in newborn mice, at a multiplicity of infection (MOI) of 1, for 1 h. At 24 hours post-infection, cells were fixed, permeabilized and stained with a mouse anti-SLEV monoclonal antibody (MSI-7, clone 6B6C-1, MAB8744; Merck Millipore, USA) followed by Alexafluor 488-labelled secondary anti-mouse (Molecular Probes, Invitrogen, USA). Experiments were performed with control group with or without primary antibodies. Stained coverslips were mounted in Mowiol 4-88 (Polysciences, Inc., USA) and analyzed using an Olympus BX61WI microscope equipped with a FV300 confocal scanning unit. Images were analyzed with imageJ software.

Identification of SLEV infection in a sick horse
From a total of 170 brain samples from horses with neurological disease received and analyzed by PCR at the Laboratório de Patologia Molecular at UFMG, one sample was identified as

Author Summary
St. Louis encephalitis virus (SLEV), a member of the Flavivirus genus, which includes West Nile encephalitis virus, Japanese encephalitis virus, Dengue virus, and other medically important viruses, is a cause of encephalitis in humans and animals. SLEV is considered endemic in the Americas, and currently there is no vaccine or specific treatment available for controlling of preventing SLEVinduced encephalitis. In this study we describe the first isolation of SLEV from an adult male horse with neurologic disease, which was further characterized by molecular and serological methods. Phylogenetic analysis of a 903 base pairs amplified sequence from partial Envelope (E) gene region indicated that the isolate from the horse was within the cluster of the VB genotype. In addition, inoculation of the SLEV isolate intracranially in newborn mice resulted in circulatory and neurological changes. This is the first report of isolation of SLEV from a horse with neurological disease in Brazil.
positive for SLEV. This brain sample was obtained in March 2009 (late summer) from a 12 years-old male horse of undefined breed, which died 72 hours after presenting neurological signs. Those neurological signs were described as incoordination, depression, and flaccid paralysis of the hind limbs. The horse came from a farm in Abaeté, countryside of Minas Gerais State, 207 kilometers from the capital, Belo Horizonte. In the same farm there were two additional horses that remained clinically healthy. Importantly, this SLEV-positive brain sample was negative for other pathogens commonly associated with encephalitis in horses, including Rabies virus, Equine Herpesvirus-1, Equine Herpesvirus-4, West Nile virus, Eastern equine encephalitis virus, Western equine encephalitis virus, Venezuelan encephalitis virus, and Sarcocystis neurona.
The SLEV amplicon originated by the RT-PCR reaction was sequenced and deposited in GenBank (Submission ID #1663732), referring to SLEV strain MG150.

Virus isolation and characterization
After confirmation of the virus identity by sequencing, we focused our efforts on isolating SLEV from the horse tissue. A brain fragment from the SLEV-positive horse was homogenized and inoculated in C6/36 mosquito cells, which is a cell lineage suitable for arbovirus propagation. All passages were tested for SLEV by RT-PCR, and were all positive beginning at the second passage. The virus was isolated after three passages, obtaining a SLEV stock of 1610 3 FFU/mL of supernatant.
To characterize this newly isolated strain in vivo, we performed serial passages of SLEV by intracranial inoculation of newborn mice, which allowed us to gather some data on MG150 strain pathogenicity. Increase of clinical signs and circulatory changes were associated with increased mortality that reached 100% at the 7 th passage (Figure 2). At third passage, edema and necrosis at the distal extremity of the hind limbs and at the tip of the tail were observed in SLEV infected newborn mice (Figure 3). After the fourth passage SLEV infection was associated with behavioral changes ranging from excitability to apathy and neurological changes including tremors, loss of proprioception, and walking in circles, which were accompanied by the same circulatory changes as observed at the third passage ( Figure 3). Importantly, neurological changes in the final SLEV passages were accompanied by hemorrhage in the CNS and peritoneum ( Table 2). Neither mortality nor clinical signs were observed in uninfected control mice.
Histological analysis indicated hyperemia and discrete multifocal hemorrhage in CNS from all infected newborn mice at the 4 th passage. Multifocal mild lympho-hystiocytic inflammatory infiltrate in the leptomeninges and around blood vessels in the brain was observed in one newborn mouse at the 4 th passage ( Figure 4A). Focal mild neuronal degeneration, and multifocal hemorrhage were also observed ( Figure 4B). Circulatory changes such as hyperemia and multifocal hemorrhage were noticed in several organs, including brain, kidney, liver, lung, and heart at the 5 th passage ( Figure 4C). Newborn mice at the 6 th passage also developed hyperemia in the kidney and lung, and multifocal hemorrhage in liver and brain ( Figure 4D). Newborn mice at the 7 th passage had hyperemia in the kidney, liver, brain, and lung ( Figure 4E). No histological changes were detected in uninfected control mice. Together, these data indicates that SLEV can cause disease in newborn mice. Virus adaptation to the murine host increased after each passage, resulting in neurological and circulatory changes consistent with Flavivirus infection, providing further evidence that our isolated virus strain was indeed SLEV.
Inocula (i.e. CNS tissue homogenate pools) resulting from each passage in mice were submitted for RNA extraction for confirmation of viral detection by RT-PCR (data not shown). The same procedure was performed with organs that had gross changes. Pool of organs, including the liver, heart, kidney, and lung from the 4 th to the 7 th passage were analyzed. CNS and other organs were positive in all RT-PCR assays, confirming the presence of viral RNA in tissues.

Detection of SLEV proteins in cell culture
To further confirm SLEV identity, we performed an immunofluorescence assay, using a commercial monoclonal antibody to detect SLEV proteins in cell culture. Monolayers of C6/36 mosquito cells were infected with the 7 th SLEV passage, fixed and stained 24 hours post-infection. Infected cells were positively stained with the anti-SLEV antibody, which was not observed in mock-infected cells, confirming the identity of this SLEV strain in an antibody-based test (data not shown).

Discussion
The first isolation of SLEV from a horse that died due to a neurological disease in Brazil is a significant event. Virus isolation from horse brain tissue, together with molecular and immunofluorescence data, confirms that SLEV was the agent that caused disease and, ultimately, the horse death in this case. To our knowledge, this is the first observation that SLEV can cause disease in wild or domestic animals, which indicate that some aspects of SLEV viral cycle and its ability to cause disease need further studies. Furthermore, a model of newborn mice infection for characterization of SLEV was thoroughly described.
In our efforts to characterize SLEV strain MG150 in vivo, to study its pathogenicity, we noticed the virus progressively adapted to serial passagens in newborn mice. The mouse has been previously used as a model for assessing SLEV virulence [41]. Disease presented by newborn mice infected with the last SLEV passages had some similarities to human disease, such as the development of hemorrhagic manifestations [19], mortality, and neurological changes [12,14]. Importantly, lesions in organs such as the liver and heart are likely to reflect a systemic circulatory change rather than any specific viral tropism for these organs.