Research Article: A novel Zika virus mouse model reveals strain specific differences in virus pathogenesis and host inflammatory immune responses

Date Published: March 9, 2017

Publisher: Public Library of Science

Author(s): Shashank Tripathi, Vinod R. M. T. Balasubramaniam, Julia A. Brown, Ignacio Mena, Alesha Grant, Susana V. Bardina, Kevin Maringer, Megan C. Schwarz, Ana M. Maestre, Marion Sourisseau, Randy A. Albrecht, Florian Krammer, Matthew J. Evans, Ana Fernandez-Sesma, Jean K. Lim, Adolfo García-Sastre, Ted C. Pierson.


Zika virus (ZIKV) is a mosquito borne flavivirus, which was a neglected tropical pathogen until it emerged and spread across the Pacific Area and the Americas, causing large human outbreaks associated with fetal abnormalities and neurological disease in adults. The factors that contributed to the emergence, spread and change in pathogenesis of ZIKV are not understood. We previously reported that ZIKV evades cellular antiviral responses by targeting STAT2 for degradation in human cells. In this study, we demonstrate that Stat2-/- mice are highly susceptible to ZIKV infection, recapitulate virus spread to the central nervous system (CNS), gonads and other visceral organs, and display neurological symptoms. Further, we exploit this model to compare ZIKV pathogenesis caused by a panel of ZIKV strains of a range of spatiotemporal history of isolation and representing African and Asian lineages. We observed that African ZIKV strains induce short episodes of severe neurological symptoms followed by lethality. In comparison, Asian strains manifest prolonged signs of neuronal malfunctions, occasionally causing death of the Stat2-/- mice. African ZIKV strains induced higher levels of inflammatory cytokines and markers associated with cellular infiltration in the infected brain in mice, which may explain exacerbated pathogenesis in comparison to those of the Asian lineage. Interestingly, viral RNA levels in different organs did not correlate with the pathogenicity of the different strains. Taken together, we have established a new murine model that supports ZIKV infection and demonstrate its utility in highlighting intrinsic differences in the inflammatory response induced by different ZIKV strains leading to severity of disease. This study paves the way for the future interrogation of strain-specific changes in the ZIKV genome and their contribution to viral pathogenesis.

Partial Text

The genus Flavivirus of the family Flaviviridae includes important vector-borne human pathogens such as dengue (DENV), yellow fever (YFV), West Nile (WNV), Japanese encephalitis (JEV), and tick-borne encephalitis viruses (TBEV). Many of these viruses are prevalent in the equatorial region of the globe that harbors the arthropod vectors required for their transmission [1–3]. Zika virus (ZIKV) is a recently emerged member of the flaviviruses which has caused large outbreaks in human populations over the last decade and was recognized as a Public Health Emergency of International Concern (PHEIC) by the World Health Organization (WHO) in February 2016. It is a mosquito borne, enveloped virus with a 10.7 kb positive sense single stranded RNA genome. Like other flaviviruses, the ZIKV genome encodes a polyprotein, which is post-translationally processed by cellular and viral proteases into three structural proteins (Capsid/C; pre-membrane/prM; Envelope/E) and seven non-structural proteins (NS1, NS2A, NS2B, NS3, NS4A, NS4B and NS5). The structural proteins protect the genome, participate in virus entry into and exit from the host cell, and are primary targets of the host antibody mediated immune response. Non-structural proteins are required for viral genome transcription and replication, proteolytic processing of the polyprotein and inhibition of cellular innate immune response [1–3]. ZIKV was first isolated in 1947 from a febrile rhesus macaque in the Ziika forest in Uganda, where it circulated between nonhuman primates and sylvatic Aedes africanus mosquitoes [4]. Subsequently, it was sporadically diagnosed in Africa and Southeast Asia as a cause of mild self-limiting febrile disease in humans. In 2007, it caused the first large-scale outbreak in the Yap islands, where more than 7,000 individuals were estimated to be infected [3]. In 2013, ZIKV appeared in French Polynesia, where it infected approximately 28,000 individuals [1–3]. Here for the first time, ZIKV infections were found to be associated with increased incidences of Guillain-Barré syndrome, a debilitating neuronal disease in adults. Also for the first time, ZIKV presence was detected in body fluids other than blood (including semen, saliva, urine), and risk of direct human-to-human transmission was observed. In 2014, ZIKV appeared in Brazil and by late 2015 the virus had spread across South and Central America, the Caribbean and the southern parts of the United States [1–3]. Importantly, during spread of ZIKV pandemic in Brazil, infections during pregnancy were linked to fetal malformations such as spontaneous abortion, stillbirth, hydrocephaly and microcephaly, and placental insufficiency to miscarriage, now collectively known as congenital Zika virus syndrome (CZVS) [5].

In November 2016 WHO declared that ZIKV is no longer a public health emergency, however it remains a serious public health concern, especially in the light of recent studies showing fetal microcephaly and other long-term cognitive disorders resulting from ZIKV infection [35]. From a scientific perspective, there is still an extensive knowledge gap on ZIKV evolution, biology and pathogenesis. From the first isolation of ZIKV in 1947 until 2015, there were only three studies which tested the virus’ pathogenic potential in animal models [12, 13, 36]. Since then there have been several more which have utilized various mouse models to study different aspect of ZIKV pathogenesis and revealed critical evidence of neuropathology and fetal abnormalities associated with ZIKV infection [8]. Mouse models remain an attractive option because of their amenability for genetic manipulation and the availability of a vast repertoire of mouse strains, which allows investigation of the role of specific host genetic loci in viral pathogenesis. Using mouse strains deficient in various components of antiviral signaling, researchers had shown a critical role of type I IFN signaling in preventing ZIKV infection in mice [14, 17, 18]. Shannan L Rossi et al compared susceptibility of A129 (lack type I IFN response) and AG129 (lack type I and II IFN response) mice to ZIKV challenge and found the latter to be more prone to severe disease [19]. This highlighted the role of type I and II IFN response against ZIKV. Later on irf3-/-irf5-/-irf7-/- triple knock out mice which are compromised in type I, II and III IFN induction were shown to be more susceptible to ZIKV induced disease as compared to Ifnar1-/- mice (lack type I IFN response) [17], which pointed to importance of type II and III IFN response against ZIKV. In our previous studies on the role of ZIKV NS5 protein in antagonizing the antiviral host response, we found that, similar to DENV, ZIKV targets STAT2 for degradation in human cells and not in mouse cells, although probably through the involvement of different host co-factors [29]. This knowledge led us to hypothesize that STAT2 could be critical for preventing ZIKV infection in animal models. Indeed in our study, Stat2-/- mice were highly susceptible to ZIKV infection and disease. In comparison Ifnar1-/- mice showed in general delayed disease symptoms and less body weight loss due to ZIKV infection than Stat2-/- mice. While Ifnar1-/- mice lack type I IFN signaling, Stat2-/- mice lack both type I and type III IFN signaling [31]. This may explains the difference in their susceptibility to ZIKV induced disease and suggests a critical role of type III IFN in mediating antiviral responses against ZIKV. This is in accordance with the findings of Bayer et al [10] who showed protective role of type III IFN against ZIKV infection in human placental tissue. However, for the differences observed between Stat2-/- and Ifnar1-/- mice, we cannot exclude the possible role of STAT2 independent, IFNAR1 dependent, type I IFN signaling. Immunocompromised mouse models are useful options when WT mice are refractory to viral infection. In our study Stat2-/- mice recapitulated ZIKV tissue tropism to CNS and gonads and displayed neurological symptoms of various degrees, which are consistent with human disease. It will be interesting to test ZIKV infection of Stat2-/- as well as Stat2-/+ heterozygous mice in pregnancy models to see whether they can simulate the transplacental infection and neuroteratogenic properties of ZIKV.