Table 2.

Overview of studies showing heterologous acute sequential or chronic co-infection can alter influenza virus immunity.

Infection type	Species	Priming strain (vaccine or infection)	Secondary strain (vaccine or infection)	Experimental design	Disease outcome	Effects on magnitude or quality of immune response	Reference
Acute, sequential (heterosubtypic)	C57BL/6	Influenza virus (H9N2 and H1N1)	Influenza virus (H7N9)	Mice were primed with 104 TCID50 of H9N2 or 102 TCID50 of H1N1 intranasally, and challenged with H7N9 intranasally at 10–12 weeks post priming	Mice primed with H9N2 or H1N1 showed increased survival, enhanced viral clearance, and decreased weight loss compared with naive mice	Prior infection with H9N2 or H1N1 leads to early and robust CD8 T cell responses during secondary infection with an antigenically distinct influenza virus, H7N9. Importantly, the magnitude of the priming-virus memory CD8 T cells was the best correlate of protection against H7N9 challenge. In addition, the degree of conferred protection (i.e., viral clearance, weight loss profile, and survival) and immunodominance of CD8 T cell responses varied by the priming-virus strain.	(52)
Acute, sequential	BALB/c	Influenza virus (X31 and H3N2)	Respiratory syncytial virus (RSV, A2 strain); recombinant Vaccinia virus (VV) expressing RSV attachment protein (rVV-G) or control β-galactosidase (rVV-β-gal)	Mice were infected with 3 × 106 PFU human RSV, 50 hemagglutinin (HA) units of X31, or HEp-2 lysate intranasally at day 0. Three to five weeks later, they were infected with 3 × 106 PFU rVV-G or rVV-β-gal via scarification, and 14 days later they were challenged with 3 × 106 PFU human RSV intranasally	Mice previously infected (Flu-G-RSV or RSV-G-RSV) exhibit decreased eosinophilia and weight loss (compared with Hep-2-G-RSV)	Flu-G-RSV mice had decreased TNF-α and IL-4 cytokine levels. In addition, 16.9 ± 2.7% of CD8 T cells recruited into the lung (post RSV infection) bound influenza tetramer, and 39.4 ± 3.8% expressed IFN-γ. Transfer of splenocytes at 21 or 149 days post influenza virus infection, followed by rVV-G and RSV challenge 14 days later also resulted in decreased eosinophilia	(35)
Acute, sequential	C57BL/6, B6.Pl-Thy1a/Cy (Thy1.1) and B6.SJLptprcapep3b/BoyJ (CD45.1)	Sendai virus (enders strain)	Influenza virus (X31 and H3N2)	Mice were infected with 250 EID50 Sendai virus and challenged 30–35 days later with 300 EID50 X31. For reverse order, mice were infected with 300 EID50 of X31 and challenged with 250 EID50 of Sendai virus 30–35 days post flu infection	Requires further investigation	Early infiltration and ~5× increase in cell number of Sendai virus specific CD8 T cells into the lungs of flu infected mice (day 4 post flu). Flu specific [nucleoprotein (NP) and polymerase acidic protein (PA)] CD8 T cell responses were unaltered, and early recruitment of memory cells was from migration of cells from other anatomical sites. When the sequence of infection was reversed, early infiltration and ~4× increase of flu specific memory CD8s occurred at day 4 post Sendai virus infection	(46)
Acute, sequential	Human	Influenza virus	Acute Epstein–Barr virus (EBV)	Influenza A virus-immune patients with acute EBV infection were recruited from the Umass Student Health Services. Age ranged from 18 to 23 years old. Acute EBV infection was confirmed using a monospot test and detection of anti-EBV capsid IgM in patient sera. Healthy volunteers were recruited from UMass Medical School. Age ranged from 24 to 50 years old	See Ref. (61)	Identified cross-reactive CD8 T cells specific for influenza A virus M158 and EBV-BMLF1280, despite only 33% sequence homology	(59)
Acute, sequential	Human	Influenza virus	Acute EBV	Influenza A virus-immune patients with acute EBV infection were recruited from the Umass Student Health Services. Age ranged from 18 to 23 years old. Acute EBV infection was confirmed using a monospot test and detection of anti-EBV capsid IgM in patient sera. Healthy volunteers were recruited from UMass Medical School. Age ranged from 42 to 50 years old	See Ref. (61)	Cross-reactive M1 and BMLF-specific CD8 T cells utilize unique clones not found in single M1 or BMLF-specific CD8 T cell pools. Computer simulation suggests the effects of cross-reactivity on T cell receptor (TCR) repertoire diversity depends on the degree of similarity between epitopes. If epitopes are structurally similar, termed “near cross-reactive,” responses will lead to a more narrow TCR repertoire, whereas cross-reactive responses between structurally divergent epitopes, termed “far cross-reactive,” will lead to a broad TCR repertoire	(60)
Acute, sequential	Human	Influenza virus	Acute EBV	College students with symptoms of acute infectious mononucleosis (AIM) were recruited. Age ranged from 18 to 30 years old. Acute EBV infection was confirmed by a monospot test and the detection of anti-EBV capsid IgM in patient serum. Healthy, EBV-seropositive donors, age >18 years old, were used as controls	CD8 T cells cross-reactive for influenza and EBV epitopes may contribute to AIM disease severity by augmenting CD8 T cell responses.	IAV-M1+/EBV-BMLF+ double positive CD8 T cells had the strongest correlation with AIM disease severity and predict severe AIM in a relative-risk analysis. Single IAV-M1 and EBV-BMLF each had weaker associations and no other tetramer + population tested (2 two from CMV and EBV) were correlated with AIM severity	(61)
Acute, sequential and chronic co-infection	C57BL/6	Influenza virus	Acute and chronic murine herpesvirus 68 (MHV68)	Mice were primed with 107.9 EID50 PR8 intraperitoneally, challenged with 106.5 EID50 X31 intranasally, and later were or were not infected with 104 PFU of MHV68 intranasally. In another study, mice were infected intranasally with MHV68, boosted intraperitoneally with 5 × 107 PFU of recombinant VV expressing MHV86 p56 peptide AGPHNDMEI (Vacc-p56), and then were or were not challenged with X31 intranasally. Each infection was delivered 6 weeks apart	Requires further investigation	Co-infected mice (PR8-X31-MHV68) show attrition of influenza (NP) and MHV68 (p79)-specific memory CD8 T cells compared with their respective single infected counterparts at day 100. The presence and degree of attrition varies by anatomical site in both cases. In addition, mice primed with MHV68 then sequentially infected with influenza virus (MHV68-vacc-p56-X31) exhibit higher numbers of influenza-specific CD8 T cells at day 14, but a lower number at day 200	(79)
Chronic co-infection	BALB/c	MHV68 (WUMS strain)	Influenza virus (PR8 and H1N1)	Mice were infected with 4 × 104 PFU MHV68 or PBS (mock-infected) and 28, 60, or 120 days later were challenged with 1 × 104 PFU PR8	Latent MHV68 infection confers protection against influenza virus challenge, as determined by improved survival, enhanced influenza viral clearance, and decreased lung injury	Co-infected mice exhibit increased levels of IFN-γ, TNF-α, and IL-12p40, but decreased levels of neutrophil chemokines CXCL1 (KC) and CXCL2 (MIP-2α). Co-infected mice also had increased numbers of CD69+ CD4+ (day 0 and 4) and CD8+ T (day 0, 4, and 6) cells in the lung, decreased neutrophils (day 8), and enhanced activation of alveolar macrophages. Adoptive transfer of macrophages from co-infected mice was sufficient to confer protection against influenza virus challenge	(47)
Chronic co-infection	C57BL/6 and IFN-γ KO	Cytomegalovirus (Smith)	Influenza virus (X31 and H3N2)	Mice were infected with 4 × 104 PFU of murine cytomegalovirus (MCMV) intraperitoneally. Co-infected mice were also infected with 1 × 106 EID50 of X31 at 5 weeks (early latency), 12 weeks (established latency), or 9 months (long-standing latency) post MCMV infection	MCMV co-infection confers protection against influenza virus challenge, but protection wanes with time and is not observed in long-standing latent MCMV infection	Mice co-infected with influenza virus at 5 or 12 weeks post MCMV infection exhibit higher influenza-specific CD8 T cell responses against three immunodominant influenza epitopes (polymerase basic protein 1, PA, and NP) and decreased influenza virus titers	(21)
Chronic co-infection	Human	Cytomegalovirus	Fluzone vaccine (each dose contains 15 µg of HA from H1N1, H3N2, and B strains)	Ninety-one healthy donors were enrolled at the Stanford-LPCH Vaccine Program in fall of 2008 (89 completed the study). The validation cohort consisted of 77 individuals who returned in fall of 2009, plus 37 subjects vaccinated in another study between 2010 and 2011 flu seasons	Young CMV seropositive subjects had higher antibody response to the Fluzone vaccine at 28 days and 1 year post vaccination, as compared with young, CMV seronegative subjects. However, no difference was observed in the elderly, based on CMV serostatus	Young CMV+ subjects have a broadly activated immune system compared with their CMV-counterparts. This is reflected by augmented expression of genes important for immune activation (e.g., antigen processing and presentation, NK cell cytotoxicity), increased levels of IL-13, IFN-γ, and CD8+ pSTAT1/3 in response to IL-6 stimulation. This study also found, elderly CMV+ subjects showed lower responses to IL-6, compared with young CMV+	(21)
Chronic co-infection	Human	Human cytomegalovirus (HCMV) and/or EBV	Influenza virus	Samples were collected from 50 patients [20 hepatitis B virus (HBV), 12 influenza, 12 dengue, 3 adenovirus, 3 fevers with unknown etiology] and 5 healthy volunteers attending clinics in Singapore or Italy. Diagnosis was confirmed utilizing appropriate methods for the infection within 5 days of selection. For example, influenza infections were confirmed with isolation of influenza A virus from nasal swabs	Requires further investigation	Acute infection with influenza, HBV, dengue, and adenovirus induce activation (CD38+ HLA-DR+) of HCMV- and EBV-specific CD8 T cells. In one influenza patient, 1/4 of activated CD8 T cells at onset were HCMV-specific, and influenza-specific CD8 T cells could not be detected until day 5. In addition, IL-15 preferentially activates memory CD8 T cells specific for chronic infections, augments anti-viral cytokine production with TCR stimulation, and is sufficient for spontaneous IFN-γ production	(50)