Key words

1 Introduction

West Nile virus (WNV) , a mosquito -borne neurotropic pathogen , belongs to the family of Flaviviridae, the genus Flavivirus , a group of plus-sense, single-stranded RNA viruses [1, 2]. It has become a global public health concern. The virus was originally isolated in Africa and later caused epidemics with mainly febrile illness in humans in Europe, Africa, the Middle East, and parts of Asia. In 1999, a more virulent WNV strain was introduced in North America and has since caused significant mortality in humans, horses , and birds . Additionally, it has been the cause of recent outbreaks of viral encephalitis in Europe and Australia [3, 4]. WNV infection of the central nervous system (CNS) commonly presents as encephalitis, meningitis , or acute flaccid paralysis. The overall mortality rate in persons who develop WNV neuroinvasive disease is about 10 %, although it increases significantly in the elderly and immunocompromised. Up to 50 % of WNV convalescent patients were reported to have long-term neurological sequelae or chronic kidney diseases [5–8]. Currently, licensed vaccines are not yet ready for human use, and treatment is mostly nonspecific and supportive.

WNV has been studied in various animal models [9–12]. Among them, mice are generally susceptible to WNV infection, easiest to work with, economic, and are most amenable to immunological manipulation. Thus, several laboratory murine models of WNV infection have been utilized to investigate the host factors involved in protective immunity and viral pathogenesis in humans. Systemic WNV infection has been commonly performed in young adult mice (6–10 weeks old). Following the initial subcutaneous (s.c.) or intraperitoneal (i.p.) inoculation , WNV induces a systemic infection and eventually invades the CNS. The severity and symptoms of lethal infection observed in mice mimic severe cases observed in humans (see Note 1 ). The incidence of WNV neuroinvasive disease has been reported to increase with age, with the highest incidence among persons aged over 70 years [13–15]. Systemic WNV infection in old mice (18–22 months old) has been used to define aging as a risk factor in the murine model [16, 17] (see Note 2 ). There is increasing evidence that persistent WNV infection contributes to long-term morbidity. WNV antigen or RNA has been detected in the brain and/or urine of WNV patients ranging from a few months to several years after the initial acute illness [18, 19]. We have recently demonstrated that WNV H8912, an isolate previously cultured from the urine of a persistently infected hamster, induces a similar persistent infection phenotype in mice following an i.p. inoculation and preferentially persists in mouse kidneys [20]. Thus, WNV H8912 systemic infection in mice can serve as a tool to study persistent renal infection (see Note 3 ). Cutaneous WNV infection either by mosquito feeding or intradermal (i.d.) injection, which presumably mimics natural infection in humans, is an important model for examining host immunity to vector -borne pathogens [21, 22] (see Note 4 ). Finally, direct CNS infection by intracranial (i.c.) injection of mouse brain is used to study the neurovirulence of WNV (see Note 5 ). In summary, studies from the above well established murine models will provide critical insights into WNV pathogenesis and protective immunity in humans.

2 Materials

  1. 1.

    Mice: We purchased 6- to 10-week-old C57BL/6 (B6) mice and 18- to 22-month-old B6 mice from Jackson Laboratories (Bar Harbor, ME) and the National Institute of Aging (Bethesda, MD), respectively. All animal experiments were approved by the Institution Animal Care and Use Committee.

  2. 2.

    WNV: The parental strain WNV NY385-99 (WNV NY99, [12]), a kind gift from Dr. R Tesh (UTMB, Galveston, TX), was passaged once in Vero cells and twice in C6/36 cells to make a virus stock (3.5 × 107 PFU/ml) for all acute infection studies. For persistent infection, WNV H8912, which was recovered from hamster urine 274 days post infection after three consecutive passages of a urine isolate from a chronically infected hamster, was used [23].

  3. 3.

    Phosphate-buffered saline with 5 % gelatin (PBSG): Dissolve 1.2 g of gelatin type B from bovine skin into 800 ml of PBS. Adjust the pH to 7.4. Autoclave the solution and cool the liquid to 50 °C. Aliquot to 10 ml aliquots until further use.

  4. 4.

    Syringe needles: A 1-cc tuberculin syringe with 26-G needle was used for systemic infection studies (acute or persistent, see Notes 1 – 3 ). A 0.5-ml, ultra-fine insulin syringe with 29-G needle was used in cutaneous and direct CNS injections (see Notes 4 and 5 ).

  5. 5.

    Anesthetizing agents: Isoflurane inhalation and Ketamine/xylazine injection were used for anesthesia of mice before systemic and direct CNS or cutaneous inoculations , respectively.

  6. 6.

    Facilities: Biosafety level 3 (BSL3) and animal BL3 (ABLS3) facilities were used for preparation of viral inocula and injection/housing of mice, respectively.

3 Methods

  1. 1.

    Preparation of viral inocula: WNV inocula were prepared by serial dilution of viral stocks in phosphate-buffered saline with 5 % gelatin (PBSG) at BL3 facilities and were kept on ice before injections. A final volume of 200 and 50 μl of viral inocula was delivered to each mouse for systemic and cutaneous or direct CNS injections, respectively.

  2. 2.

    Injection procedure:

    1. (a)

      All animal inoculations were performed in ABL3 facilities in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health.

    2. (b)

      For systemic infection studies, isoflurane inhalation was used first for anesthesia of mice in a small chamber. Next, the tuberculin 1 cc syringe was uncapped, the amount of viral inoculum needed was drawn up into the syringe and needle, and 200 μl of virus inoculum was injected in the left quadrant of the mouse abdomen towards the head at a 30–40° angle (see Notes 1 – 3 ).

    3. (c)

      For i.d or i.c. inoculation, we first injected 200 μl of Ketamine/xylazine i.p. (administered at 70 mg/7 mg/kg mouse weight). Next, we used a 0.5-ml insulin syringe with a 29 G needle to inject anesthetized mice with WNV isolate in 50 μl PBSG.

    4. (d)

      For i.d. injection, insert the needle into the dorsal side of the ear, make sure not to insert the needle too far and either go through the ear or into the skull (see Note 4 ).

    5. (e)

      For i.c. injection, carefully insert the needle where the midline of the skull intersects with the midline between the eyes and make sure to use enough force to insert the needle into the skull, but not too deep into the brain to cause damage (see Note 5 ).

  3. 3.

    Post-infection monitor:

    1. (a)

      Infected mice were housed in ABL3 facilities and monitored twice daily for 1 month.

    2. (b)

      When mice showed any sign of morbidity (obvious severe illness, unable to right self when tipped on side or back, obvious extreme weight loss or signs of paralysis), they were immediately collected and euthanized by using CO2.

  4. 4.

    Conclusions: Murine models of WNV infection have provided important insights into host immunity and viral pathogenesis in humans. Results from animal studies will help to identify therapeutic goals to modulate immune functions pharmacologically, as well as antiviral countermeasures, and to develop vaccines that can induce robust memory T cell responses.

4 Notes

  1. 1.

    Acute systemic infection in young adult mice: Following the initial i.p. infection, wild-type WNV NY99 strain was found to induce a systemic infection and eventually invaded the CNS [10, 24]. The virus quickly spread to the spleen, kidneys, and other peripheral tissues . Further, WNV crossed the blood brain barrier (BBB) after a brief viremia and entered the CNS (see Table 1). An increased viral burden in serum usually correlated with earlier viral entry into the brain [25]. Moreover, systemic WNV replication was reported to induce a pro-inflammatory cytokine response that modulates BBB permeability, which in turn, may enable viral entry into the brain and induce lethal encephalitis [26, 27]. Thus, it is critical to control virus dissemination in the periphery during early stages of WNV infection. WNV-infected mice developed symptoms starting on day 6, and succumbed to infection when encephalitis developed, usually within 1–2 weeks [10, 24, 28]. Similar outcomes have been reported following s.c. infection in young adult mice [16, 25]. The LD50 and LD100 of wild-type WNV NY99 i.p. infections in young adult mice are 100 plaque forming unit (PFU) and 1000 PFU, respectively.

    Table 1 WNV infection models in mice
    Full size table
  2. 2.

    Acute systemic infection in aged mice: Following either i.p. or s.c. WNV infection, aged mice (18–22 months old) displayed persistently elevated viremia and higher viral loads in the brain (see Table 1). As a result, aged mice were much more susceptible to systemic WNV NY99 infection compared to young adult mice (6–10 weeks old). The decline in immunity seen in the elderly is a significant contributor to the increased risk of infection. Impaired innate and adaptive T cell functions in aged mice contribute to an enhanced host susceptibility to WNV encephalitis [16, 17]. Toll-like receptor (TLR)7 provides co-stimulatory signals for γδ T cell activation during WNV infection. An impaired TLR7 signaling led to dysfunction of these cells observed in WNV-infected aged mice [29].

  3. 3.

    Persistent WNV infection in young adult mice: WNV antigen or RNA has been detected in the brain or urine of WNV convalescent patients, ranging from a few months to several years after the initial acute illness [18, 19]. To define a murine model of persistent renal infection, we used WNV H8912, an isolate recovered from the urine of chronically infected hamsters. The virus showed a significantly reduced neuroinvasiveness in young adult mice. All mice survived i.p. inoculation of a dose up to 106 PFU of WNV H8912, and only about 20 % of infected mice developed mild disease symptoms. WNV H8912 induced constitutive interleukin (IL)-10 production and a delayed antiviral response. Viral RNA was detected quickly in blood and spleen but much later in kidneys (see Table 1). The virus persisted preferentially in the kidneys with mild renal inflammation, and less frequently in the spleen for up to 2.5 months post infection. This was concurrent with detectable serum WNV-specific IgM and IgG production [20]. In another study, Appler et al. [30] inoculated young adult mice by the s.c. route with a New York (2000) isolate (NY99 genotype) and for 16 months examined their tissues for infectious virus and WNV RNA. Infectious WNV persisted for 1 month in all mice and was found in 12 % of mice at 4 months; WNV RNA persisted for up to 6 months in 12 % of mice that had subclinical infections. The frequency of persistence was tissue dependent and found in the skin, spinal cord, brain, lymphoid tissues, kidneys, and heart, but the persistence less frequently in the kidney tissues [30, 31]. In summary, these two models share some similarity and discrepancy in symptoms and tissue tropism with the clinical findings in some WNV convalescent patients having long-term morbidity (either chronic kidney diseases or long-term neurological sequelae ).

  4. 4.

    Cutaneous infection model: WNV infection in mice by i.d. challenge partially mimics natural infection in humans, though it is much less complicated due to a lack of concurrent exposure to components of mosquito saliva [32, 33]. One earlier study [34] revealed that the mean and median doses of WNV inoculated by Culex tarsalis mosquitoes as they probe and feed on peripheral tissues of a mouse were 104.3 PFU and 105 PFU. The same doses of WNV by i.d. infection in mice results in survival rates of 25–40 %. Cutaneous infection shares a disease outcome that is similar to those of systemic infection. However, its initial virus dissemination in mice differs from the latter. Following i.d. WNV challenge , the virus is first deposited in the skin cells and is further disseminated to lymph nodes and other peripheral tissues by the migration of the epidermal dendritic cells-Langerhans cells (LCs) to initiate systemic infection [35, 36]. TLR7 innate signaling contributes to host defense mechanisms, which leads to reduced viremia and lethality when WNV infection of mice was initiated by i.p. injection [37]. Nevertheless, the TLR7-induced immune response in mouse skin following cutaneous inoculation plays a role in viral pathogenesis by promoting WNV dissemination from the skin to initiate systemic infection. In particular, WNV infection in epidermal keratinocytes triggers a TLR7-dependent production of IL-1β, IL-6, and IL-12, which promotes LC migration from the skin to other peripheral organs. This process compromises the protective effects of TLR7 signaling during the systemic infection stage.

  5. 5.

    Direct CNS infection: It is often used to study the neurovirulence of WNV. The LD50 of wild-type WNV NY99 following i.c. inoculation is usually lower than that of systemic infection. Wild-type WNV induces neuroinvasive disease within 2 weeks following i.c. challenge [38]. WNV H8912, which is highly attenuated in neuroinvasiveness, induces no lethality following systemic infection even with a dose up to 106 PFU. Compared to wild-type WNV, WNV H8912 also displayed a reduced neurovirulence in mice by i.c. challenge. Mice inoculated with 104 and 106 PFU of WNV H8912 succumbed to infection; while 41 % of mice inoculated i.c. with 102 PFU of WNV H8912 survived [20].