ABSTRACT
West Nile virus (WNV) infection outcomes vary among individuals, with most infections resulting in asymptomatic or mild flu-like symptoms. We previously reported an association between early cytokine production and symptom outcome following WNV infection in US blood donors. In this meta-analysis, we found that WNV-infected females reported more symptoms than WNV-infected males, despite similar initial viremia and type I interferon responses. As the infection progressed, males exhibited a protracted cytokine response—marked by sustained CCL2 (MCP-1), CCL11 (eotaxin-1), CXCL10 (IP-10) and IL-15—that was absent in females. Our results suggest that sex differences may be a factor in sustaining WNV immunity.
Keywords: blood donations, clinical outcome, blood bank, susceptibility, sex differences, cytokines
Sex may play a role in regulating the host cytokine response to West Nile virus infection in humans.
ABBREVIATIONS
- WNV
West Nile virus
- IL-15
interleukin 15
- IP-10
interferon γ-induced protein 10
- MCP-1
monocyte chemoattractant protein 1
BACKGROUND
First detected in the United States in 1999, West Nile virus (WNV) has become a major cause of encephalitis and meningitis, with annual outbreaks reported in nearly every state (Gubler 2007; Curren et al. 2018). While less than 1% of infected individuals develop neuroinvasive disease, about 20% of infections present as West Nile fever, characterized by flu-like symptoms including fever, headache and myalgia (Debiasi and Tyler 2006). Currently, there are no specific therapies or human vaccines available to treat and prevent WNV infection, and little is known about the host factors affecting the immune response to WNV (Debiasi and Tyler 2006). One such biological variable that should be considered in immunological studies is sex. Many studies observe that the immune response varies between males and females following viral infection (Klein 2012). During infection, females have been observed to mount stronger humoral and adaptive immune responses than males (Klein 2012). Prior studies have identified a heightened antiviral response among females in response to hepatitis B virus, HIV and Coxsackie virus, but this may come at the cost of increased inflammation, as evidenced by both higher rates of autoimmune disease and increased fatality following influenza A virus infection (Klein 2012; Ghosh and Klein 2017). However, reports on the sex-specific regulation of the human cytokine response to vector-transmitted viruses, such as flaviviruses and alphaviruses, are lacking.
In a previous study, we identified immune factors upregulated in response to WNV infection, including the intensity and timing of cytokines functioning in the inflammatory immune response (Hoffman et al. 2016). Given the growing body of literature reporting sex differences in the immune response, we re-evaluated these data to investigate the extent to which sex differences impact immunity to WNV. In this meta-analysis, we evaluated the same cohort of 115 WNV-infected individuals (WNV-positive cases) and 31 WNV-uninfected individuals (WNV-negative controls) and stratified these populations on the basis of sex. Symptoms that developed in the two-week time frame following blood donation were recorded for all individuals via a standardized questionnaire. Measured plasma protein levels of a panel of cytokines and chemokines, as previously reported, were re-analyzed to consider the predictive value of these proteins with regard to symptom development in the context of sex.
METHODS
Identification and characterization of WNV-infected and uninfected blood donors
All blood donations collected by the American Red Cross between June 2008 and December 2011 (n = 24,224,786) were screened for WNV RNA reactivity as previously described (Hoffman et al. 2016). WNV-reactive donors were contacted to obtain informed consent for enrollment/participation in follow-up studies in order to confirm WNV immunoglobulin M (IgM) seroconversion, deliver a questionnaire regarding symptoms and administer subsequent qualitative and quantitative WNV RNA testing (National Genetics Institute). As previously reported, subsequent WNV RNA and IgM seroconversion tests were used to identify 115 true-positive and 31 false-positive blood donors, referred to as ‘WNV-positive cases’ and ‘WNV-negative controls’, respectively. Prior to learning their true WNV infection status, all donors were administered a follow-up, standardized check-box questionnaire to determine symptoms that developed since the initial donation as previously described (Hoffman et al. 2016). Information was collected for age, race, sex, date of blood sample collection and the presence or absence of 15 listed symptoms consistent with WNV infection and other self-reported symptoms during the two weeks following donation. The 15 listed symptoms were fever, chills, headache, painful eyes, severe muscle pain, swollen glands, new rash, seizures, tremors, new difficulty with thinking, generalized weakness, joint pain, bone pain, vomiting or diarrhea and abdominal pain. Plasma cytokines and chemokines were quantified in our previous study and re-analyzed in this meta-analysis (Hoffman et al. 2016).
Statistical analysis
Scatter plots and linear regression analyses were conducted to identify relationships between continuous and discrete variables using Prism GraphPad Software, version 7.0 (San Diego, California). The two-sided Mann–Whitney U test was used to calculate the statistical significance of single comparisons of the cytokine/chemokine levels, viral load and symptom number between sexes within each WNV infection group or between WNV-positive cases and WNV-negative controls within each sex. A Holm–Šídák test was performed on cytokine/chemokine level comparisons between WNV-negative controls, IgM-negative cases and IgM-positive cases within each sex to adjust the significance level for multiple comparisons.
RESULTS
A meta-analysis was performed on data collected from the plasma samples of 115 WNV-positive cases and 31 WNV-negative controls previously reported (Table 1). We first compared the total number of self-reported symptoms between male and female donors and found that among the WNV-positive cases, females reported significantly more symptoms than males (P = 0.0002; Fig. 1A). In fact, when comparing symptom reporting between controls and WNV-positive cases, females reported significantly more symptoms following WNV infection (P < 0.0001), but males did not (P = 0.7824). While there may be sex differences in symptom reporting during viral infection, males reported significantly more symptoms than females among WNV-negative controls (P = 0.0469). To determine whether the increased symptom reporting among females could be attributed to greater viral infection, we evaluated viremia among donors based on sex (Fig. 1B). No difference in plasma viral load was observed between sexes among the WNV-positive blood donors (P = 0.1228), indicating that the increased number of symptoms experienced following WNV infection was not due to increased viral replication. Similarly, induction of the innate immune response as measured by interferon α (IFN-α) production revealed no sex differences among the WNV-negative controls ( P = 0.4516) or WNV-positive cases (P = 0.2530; Fig. 1C). These results show that despite having similar viremia and innate immune induction following WNV infection, WNV-infected females experienced more symptoms compared to males.
Table 1.
Characteristics of WNV-positive cases and WNV-negative controls by sex.
| Females | Males | P valuea | ||||||
|---|---|---|---|---|---|---|---|---|
| Characteristic | Group A: Controls (n = 14) | Group B: Cases (n = 47) | Group C: Controls (n = 17) | Group D: Cases (n = 68) | A vs B | C vs D | A vs C | B vs D |
| Symptoms, total number, mean ± SD | 0.1 ± 0.3 | 3.1 ± 3.1 | 1.2 ± 2.3 | 1.1 ± 1.5 | <0.0001 | 0.7824 | 0.0469 | 0.0002 |
| Viral Load, copies/mL, median (IQR) | — | 310 (50–8100) | — | 100 (5–1350) | — | — | — | 0.1228 |
| IFN-α, pg/mL, median (IQR) | 0.0215 (0.0215–0.0215) | 0.0215 (0.0215–0.0215) | 0.0215 (0.0215–11.8) | 0.0215 (0.0215–3.85) | 0.0045 | 0.0018 | 0.4516 | 0.2530 |
Abbreviations: IFN-α, interferon α; IQR, interquartile range; SD, standard deviation; WNV, West Nile virus.
Determined by the Mann–Whitney U test. Values <0.05 are considered statistically significant.
Figure 1.
Females report more symptoms following WNV infection. Plotted is the total number of self-reported symptoms experienced during the two weeks post-donation by WNV-positive and WNV-negative blood donors (i.e. cases and controls) stratified by sex, with females as red circles and males as blue triangles (A). Plasma WNV RNA load was quantified at the time of donation among WNV-positive cases and stratified by sex (B). Plasma IFN-α levels were measured in male and female donors among cases and controls (C). Bars indicate median values with interquartile range. *P < 0.05, ***P < 0.001 and ****P < 0.0001, by the Mann–Whitney U test. Abbreviations: IFN-α, interferon α; ns, not significant; WNV, West Nile virus.
Since viremia and IFN-α production did not explain the trends in the number of self-reported systems, we next analyzed interferon signatures to elucidate sex differences in symptom reporting among the cases and controls. WNV-positive cases were stratified temporally on the basis of IgM seropositivity, with IgM-negative cases (pre-IgM) having been infected more recently than IgM-positive cases (post-IgM). The levels of all previously measured cytokines and chemokines were compared by sex between WNV-positive IgM-negative cases, WNV-positive IgM-positive cases and WNV-negative controls (Table 2) as well as between sexes within the case and control groups (Table 3). We previously reported that individuals who reported fewer symptoms following WNV infection had elevated cytokine levels in the post-IgM phase of WNV infection (Hoffman et al. 2016). Fig. 2A–D shows that the levels of monocyte-associated chemokine CCL2 (monocyte chemoattractant protein 1 [MCP-1]), eosinophil chemoattractant CCL11 (eotaxin-1), T cell chemoattractant CXCL10 (interferon γ-induced protein 10 [IP-10]) and innate immunity factor IL-15 were significantly elevated among WNV-infected males in the post-IgM phase compared to WNV-uninfected control males ( P = 0.0298, 0.0114, 0.0076, 0.0048, respectively); this difference was not observed among females (P = 0.9056, 0.8687, 0.2713, 0.7723, respectively). In addition, the WNV-infected male blood donors had higher levels of CCL2 (MCP-1), CCL11 (eotaxin-1), CXCL10 (IP-10) and IL-15 than the female blood donors in the post-IgM phase (P = 0.0278, 0.0014, 0.0021, 0.0099, respectively). In fact, CCL11 (eotaxin-1) levels were significantly elevated in males compared to females as early as the pre-IgM phase (P = 0.0176). Interestingly, CCL11 (eotaxin-1) levels were positively correlated with age among WNV-infected males, but not among females (P = 0.0047, 0.2974; Fig. S1, Supporting Information). Altogether, these data show that males and females initially have a similar response to WNV infection, but as the infection progresses, sex differences manifest in the immune response that may impact the outcome of infection.
Table 2.
Cytokine and chemokine levels among WNV-positive cases and WNV-negative controls by sex.
| Level, pg/mL, Median (IQR) | P valuea | ||||||
|---|---|---|---|---|---|---|---|
| Cytokine/Chemokineb | Group A: Controls (n1 = 14, n2 = 17)c | Group B: IgM-negative cases (n1 = 32, n2 = 32)c | Group C: IgM-positive cases (n1 = 15, n2 = 36)c | A vs B | B vs C | A vs C | |
| Females | CCL2 (MCP-1) | 29.99 (19.97–35.24) | 42.22 (25.81–53.73) | 31.67 (20.92–34.11) | 0.0539 | 0.0288 | 0.9056 |
| CCL7 (MCP-3) | 0.14 (0.14–1.813) | 1.04 (0.14–4.87) | 0.14 (0.14–1.42) | 0.3141 | 0.4115 | 0.9750 | |
| CCL8 (MCP-2) | 13.56 (6.345–21.63) | 26.98 (15.68–43.88) | 12.22 (5.54–14.62) | 0.0207 | 0.0012 | 0.5831 | |
| CCL11 (eotaxin-1) | 12.63 (3.71–57.37) | 23.08 (3.71–76.06) | 26.13 (3.71–74.12) | 0.9087 | 0.7484 | 0.8687 | |
| CCL13 (MCP-4) | 6.17 (6.17–52.89) | 78.99 (11.67–119.5) | 67.21 (6.17–102) | 0.0332 | 0.4919 | 0.1381 | |
| CCL20 (MIP-3α) | 1.89 (1.89–1.89) | 5.74 (1.89–12.31) | 4.68 (1.89–13.7) | 0.0054 | 0.9395 | 0.0076 | |
| CXCL9 (MIG) | 144.5 (109.8–266.8) | 218.5 (163.5–298.8) | 130 (105–218) | 0.3143 | 0.1233 | 0.5306 | |
| CXCL10 (IP-10) | 35.68 (10.17–193.5) | 228 (87.97–294.5) | 13.11 (2.05–32.91) | 0.0197 | <0.0001 | 0.2713 | |
| IFN-α | 0.0215 (0.0215–0.0215) | 4.691 (0.0215–18.44) | 0.0215 (0.0215–0.0215) | <0.0001 | <0.0001 | 0.4828 | |
| IL-15 | 0.43 (0.195–0.6425) | 0.64 (0.43–1.01) | 0.36 (0.18–0.52) | 0.0918 | 0.0501 | 0.7723 | |
| IL-1RA | 69.93 (19.39–108.8) | 57.43 (34.8–92.42) | 32.94 (12.5–77.79) | 0.6758 | 0.3655 | 0.3434 | |
| MMP-9 | 1328 (901.5–3376) | 1532 (952.8–3033) | 2695 (1960–3750) | 0.9155 | 0.0391 | 0.5059 | |
| Males | CCL2 (MCP-1) | 32.04 (11.27–40.53) | 54.48 (29.67–63.7) | 41.92 (33.76–53.48) | 0.0075 | 0.0607 | 0.0298 |
| CCL7 (MCP-3) | 5.26 (0.14–16.48) | 2.105 (0.14–6.32) | 0.14 (0.14–1.838) | 0.2232 | 0.0339 | 0.0102 | |
| CCL8 (MCP-2) | 9.81 (5.215–12.83) | 26.87 (15.33–45.49) | 14.62 (10.11–21.55) | 0.0020 | <0.0001 | 0.0059 | |
| CCL11 (eotaxin-1) | 46.03 (3.71–82.89) | 72.46 (3.71–164.5) | 99.69 (33.84–165.5) | 0.1510 | 0.5538 | 0.0114 | |
| CCL13 (MCP-4) | 51.86 (6.17–101.4) | 68.13 (6.17–132.3) | 89.12 (40.17–125.3) | 0.6536 | 0.7061 | 0.5262 | |
| CCL20 (MIP-3α) | 1.89 (1.89–7.885) | 2.69 (1.89–11.27) | 2.63 (1.89–14.06) | 0.7226 | 0.6612 | 0.6392 | |
| CXCL9 (MIG) | 100 (31.71–191.5) | 167 (105–267) | 183.5 (134.3–244.8) | 0.2625 | 0.3672 | 0.0335 | |
| CXCL10 (IP-10) | 3.25 (2.05–49.49) | 170.5 (47.7–281) | 44 (15.74–91.78) | <0.0001 | 0.0072 | 0.0076 | |
| IFN-α | 0.0215 (0.0215–0.0215) | 3.229 (0.0215–8.302) | 0.0215 (0.0215–0.0215) | <0.0001 | <0.0001 | 0.1634 | |
| IL-15 | 0.43 (0.18–0.54) | 0.84 (0.3725–1.135) | 0.6 (0.3775–0.9575) | 0.0009 | 0.3382 | 0.0048 | |
| IL-1RA | 7.41 (3.86–17.4) | 65.91 (26.95–96.85) | 32.23 (7.125–55.98) | <0.0001 | 0.0081 | 0.0155 | |
| MMP-9 | 2192 (1401–3216) | 1462 (913–3055) | 2050 (1100–3333) | 0.3388 | 0.3337 | 0.6849 | |
Abbreviations: IFN-α, interferon α; IgM, immunoglobulin M; IL-15, interleukin 15; IL-1RA, interleukin 1 receptor antagonist; IP-10, interferon γ-induced protein 10; IQR, interquartile range; MCP, monocyte chemoattractant protein; MIG, monokine induced by gamma interferon; MIP-3α, macrophage inflammatory protein 3α; WNV, West Nile virus.
Determined by the Mann–Whitney U test with Holm–Šídák correction. Values <0.05 are considered statistically significant.
The significance pattern of cytokine level differences among controls, pre-IgM cases and post-IgM cases were compared between sexes. Cytokines that did not show sex differences in cytokine level comparisons are not shown.
Number of females (n1) and males (n2) in each group.
Table 3.
Statistical comparisons of cytokines and chemokines among WNV-positive cases and WNV-negative controls by sex.
| Females vs males, P valuea | |||
|---|---|---|---|
| Cytokine/Chemokineb | Controls (n1 = 14, n2 = 17)c | IgM-negative cases (n1 = 32, n2 = 32)c | IgM-positive cases (n1 = 15, n2 = 36)c |
| CCL2 (MCP-1) | 0.7614 | 0.0507 | 0.0014 |
| CCL7 (MCP-3) | 0.0140 | 0.4141 | 0.8392 |
| CCL8 (MCP-2) | 0.1830 | 0.8860 | 0.1477 |
| CCL11 (eotaxin-1) | 0.4685 | 0.0176 | 0.0021 |
| CCL13 (MCP-4) | 0.0971 | 0.7802 | 0.2583 |
| CCL20 (MIP-3α) | 0.0632 | 0.3938 | 0.8764 |
| CXCL9 (MIG) | 0.1304 | 0.0745 | 0.145 |
| CXCL10 (IP-10) | 0.0444 | 0.4055 | 0.0278 |
| IFN-α | 0.4516 | 0.4628 | 0.2925 |
| IL-15 | 0.5353 | 0.3794 | 0.0099 |
| IL-1RA | 0.0005 | 0.9389 | 0.4943 |
| MMP-9 | 0.4213 | 0.7439 | 0.1297 |
Abbreviations: IFN-α, interferon α; IgM, immunoglobulin M; IL-15, interleukin 15; IL-1RA, interleukin 1 receptor antagonist; IP-10, interferon γ-induced protein 10; MCP, monocyte chemoattractant protein; MIG, monokine induced by gamma interferon; MIP-3α, macrophage inflammatory protein 3α; WNV, West Nile virus.
Determined by the Mann–Whitney U test. Values <0.05 are considered statistically significant.
The significance pattern of cytokine level differences among controls, pre-IgM cases and post-IgM cases were compared between sexes. Cytokines that did not show sex differences in cytokine level comparisons are not shown.
Number of females (n1) and males (n2) in each group.
Figure 2.
Males exhibit a prolonged cytokine response following WNV infection that is absent among females. Cytokine and chemokine levels in plasma obtained from WNV-positive and WNV-negative blood donors (i.e. cases and controls) before and after IgM seroconversion were evaluated for (A) CCL2 (MCP-1), (B) CCL11 (eotaxin-1), (C) CXCL10 (IP-10) and (D) IL-15 by sex, with females as red circles and males as blue triangles. Bars indicate median values with interquartile range. *P < 0.05 and **P < 0.01, by the Mann–Whitney U test with a Holm–Šídák post-hoc test to adjust for multiple comparisons. Abbreviations: IgM, immunoglobulin M; IL-15, interleukin 15; IP-10, interferon γ-induced protein 10; MCP-1, monocyte chemoattractant protein 1; WNV, West Nile virus.
DISCUSSION
This meta-analysis is the first to identify sex differences in cytokine production following WNV infection. Among our cohort of 115 WNV-positive blood donors, we found that WNV-infected females reported more symptoms than WNV-infected males; this difference was not driven by greater initial viral replication or innate host responses since no differences were observed in plasma viremia and interferon α (IFN-α) production. However, immune response differences were observed between WNV-infected males and females as the infection progressed, with WNV-infected males exhibiting an altered cytokine response primarily in the post-IgM phase that was marked by changes in CCL2 (MCP-1), CCL11 (eotaxin-1), CXCL10 (IP-10) and IL-15, suggesting a potential cytokine signature associated with symptomatology.
A large body of literature has emerged supporting the idea that the immune systems of males and females differ following infection, but very few studies have evaluated the early immune response to WNV with regards to sex. In a study by Zou et al., WNV-infected females were more likely to have symptomatic WNV infection despite having similar levels of plasma viremia compared to males (Zou et al. 2010). Our analysis confirms these findings and identifies cytokine correlates associated with symptom development. Interestingly in the United States, WNV-infected males have a higher rate of hospitalization and a higher incidence of neuroinvasive disease than WNV-infected females (Gubler 2007; Curren et al. 2018). Our findings, however, suggest that females are more susceptible to symptomatic WNV within our study population. This apparent discrepancy may be explained by the spectrum of infection outcomes captured by our cohort, which include individuals who were asymptomatic and subclinical, whereas neuroinvasive and/or hospitalized cases represent the most severe subset of all WNV infections.
We previously found that the elevation of CCL2 (MCP-1), CCL11 (eotaxin-1) and CXCL10 (IP-10) in the post-IgM seroconversion phase correlated with improved symptom outcomes following WNV infection among blood donors (Hoffman et al. 2016). Upon sex stratification, we observed the same cytokine trends among males but not females. Interestingly, we found that IL-15, which was not originally associated with outcome of WNV infection in our previous study, also displayed a similar pattern upon sex stratification. Functionally, CCL2 (MCP-1) mobilizes monocytes from the bone marrow and recruits monocytes/macrophages to inflamed tissues through interactions with CCR2; this mobilization and recruitment is impaired during WNV infection in Ccr2- and Ccl2-deficient mice (Lim et al. 2011; Bardina et al. 2015). CCL11 (eotaxin-1) and CXCL10 (IP-10) influence the granulocyte and T cell response through interactions with CCR3 and CXCR3, respectively, the latter of which has been shown to be protective against WNV in mice (Glass et al. 2005; Klein et al. 2005; Zhang et al. 2008). IL-15 also influences T cell proliferation in response to viral infections through the IL-15 receptor complex and has been previously been shown to function as a chemoattractant that can draw T cells directly to inflamed tissue (Badolato et al. 1997; Wilkinson and Liew 1995; Jonuleit et al. 1997). Our results implicate these monocyte and T cell-associated chemokines as key molecular regulators of WNV pathogenesis in humans. CCL11 (eotaxin-1) has also been shown to be elevated in human neurodegenerative disease and the plasma of aging humans and mice (Huber et al. 2018). Our study is the first to implicate CCL11 (eotaxin-1) as potentially protective during WNV infection. While elevated CCL11 may play a role in severe cases of age-related neurodegenerative disease, its induction may mediate better symptom outcomes among our cohort's male donors experiencing mild WNV infection.
Immune differences between males and females have been partly attributed to the sex hormones estrogen and androgen, which have been shown to influence the immune response to viral infection (Kovats 2015). Future immunological studies on cytokine production during WNV and other viral infections should evaluate levels of these hormones to analyze inflammation in the context of sex. Additionally, IFN-α production as part of the antiviral, type I IFN response has been shown to differ between males and females in vitro (Ziegler et al. 2017). While we did not observe this in our study, sex differences in IFN-α production may be found at more specific time points of infection or manifest in more severe cases of WNV infection, such as hospitalizations and neuroinvasive WNV disease.
One limitation of this study is sample size, which prevented us from further stratifying the cases into potentially more informative subsets, such as the degree of symptom reporting. A larger cohort would have provided increased statistical power that could have allowed for additional correlations to emerge, especially between cytokine production and the degree of symptomatology. Since we conducted an analysis retrospectively of a non-concurrent prospective cohort, we did not have data on sex-specific markers nor evaluate the mechanisms of sex differences in WNV disease. In future studies, plasma protein panels could be expanded to include estrogens, androgens and other factors implicated in the sex-regulated type I IFN response. Taken together, our data suggest that sex-dependent differences in the host response to WNV may be a biological factor that contributes to outcome of WNV infection. Our data highlight a pattern of cytokine production associated with these differences.
Supplementary Material
ACKNOWLEDGEMENTS
We thank Dr. Emilia Bagiella for statistical consultation.
FUNDING
This work was supported by the National Institute of Allergy and Infectious Diseases and the National Institutes of Health [R01 AI108715-01 and T32 AI007647].
Conflict of interest. None declared.
REFERENCES
- Badolato R, Ponzi AN, Millesimo Met al.. Interleukin-15 (IL-15) induces IL-8 and monocyte chemotactic protein 1 production in human monocytes. Blood. 1997;90:2804–9. [PubMed] [Google Scholar]
- Bardina SV, Michlmayr D, Hoffman KWet al.. Differential roles of chemokines CCL2 and CCL7 in monocytosis and leukocyte migration during West Nile virus infection. J Immunol. 2015;195:4306–18. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Curren EJ, Lehman J, Kolsin Jet al.. West Nile virus and other nationally notifiable arboviral diseases-United States, 2017. MMWR Morb Mortal Wkly Rep. 2018;67:1137–42. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Debiasi RL, Tyler KL. West Nile virus meningoencephalitis. Nat Clin Pract Neurol. 2006;2:264–75. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Ghosh S, Klein RS. Sex drives dimorphic immune responses to viral infections. J Immunol. 2017;198:1782–90. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Glass WG, Lim JK, Cholera Ret al.. Chemokine receptor CCR5 promotes leukocyte trafficking to the brain and survival in West Nile virus infection. J Exp Med. 2005;202:1087–98. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Gubler DJ. The continuing spread of West Nile virus in the western hemisphere. Clin Infect Dis. 2007;45:1039–46. [DOI] [PubMed] [Google Scholar]
- Hoffman KW, Sachs D, Bardina SVet al.. Differences in early cytokine production are associated with development of a greater number of symptoms following west Nile virus infection. J Infect Dis. 2016;214:634–43. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Huber AK, Giles DA, Segal BMet al.. An emerging role for eotaxins in neurodegenerative disease. Clin Immunol. 2018;189:29–33. [DOI] [PubMed] [Google Scholar]
- Jonuleit H, Wiedemann K, Muller Get al.. Induction of IL-15 messenger RNA and protein in human blood-derived dendritic cells: a role for IL-15 in attraction of T cells. J Immunol. 1997;158:2610–5. [PubMed] [Google Scholar]
- Klein RS, Lin E, Zhang Bet al.. Neuronal CXCL10 directs CD8+ T-cell recruitment and control of West Nile virus encephalitis. J Virol. 2005;79:11457–66. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Klein SL. Sex influences immune responses to viruses, and efficacy of prophylaxis and treatments for viral diseases. Bioessays. 2012;34:1050–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kovats S. Estrogen receptors regulate innate immune cells and signaling pathways. Cell Immunol. 2015;294:63–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Lim JK, Obara CJ, Rivollier Aet al.. Chemokine receptor Ccr2 is critical for monocyte accumulation and survival in West Nile virus encephalitis. J Immunol. 2011;186:471–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Wilkinson PC, Liew FY. Chemoattraction of human blood T lymphocytes by interleukin-15. J Exp Med. 1995;181:1255–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Zhang B, Chan YK, Lu Bet al.. CXCR3 mediates region-specific antiviral T cell trafficking within the central nervous system during West Nile virus encephalitis. J Immunol. 2008;180:2641–9. [DOI] [PubMed] [Google Scholar]
- Ziegler SM, Beisel C, Sutter Ket al.. Human pDCs display sex-specific differences in type I interferon subtypes and interferon alpha/beta receptor expression. Eur J Immunol. 2017;47:251–6. [DOI] [PubMed] [Google Scholar]
- Zou S, Foster GA, Dodd RYet al.. West Nile fever characteristics among viremic persons identified through blood donor screening. J Infect Dis. 2010:202:1354–61. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.