Innate and adaptive T cells in influenza disease

Abstract

Influenza is a major global health problem, causing infections of the respiratory tract, often leading to acute pneumonia, life-threatening complications and even deaths. Over the last seven decades, vaccination strategies have been utilized to protect people from complications of influenza, especially groups at high risk of severe disease. While current vaccination regimens elicit strain-specific antibody responses, they fail to generate cross-protection against seasonal, pandemic and avian viruses. Moreover, vaccines designed to generate influenza-specific T-cell responses are yet to be optimized. During natural infection, viral replication is initially controlled by innate immunity before adaptive immune responses (T cells and antibody-producing B cells) achieve viral clearance and host recovery. Adaptive T and B cells maintain immunological memory and provide protection against subsequent infections with related influenza viruses. Recent studies also shed light on the role of innate T-cells (MAIT cells, gd T cells, and NKT cells) in controlling influenza and linking innate and adaptive immune mechanisms, thus making them attractive targets for vaccination strategies. We summarize the current knowledge on influenza-specific innate MAIT and gd T cells as well as adaptive CD8⁺ and CD4⁺ T cells, and discuss how these responses can be harnessed by novel vaccine strategies to elicit cross-protective immunity against different influenza strains and subtypes.

Keywords

influenza / innate T cells / CD4 ⁺ and CD8 ⁺ T cells / vaccination

Introduction

Influenza viruses are enveloped negative-sense RNA viruses with segmented genomes from the family Orthomyxoviridae that cause a highly contagious and rapidly spreading acute respiratory infection (reviewed in Ref. 1). Annual seasonal influenza epidemics result in significant morbidity and mortality, especially among high-risk groups, including children, the elderly, pregnant women, obese individuals, individuals with co-morbidities and indigenous populations. Sporadically, a new influenza virus emerges from an animal host and spreads rapidly in the susceptible population, resulting in a pandemic often associated with significant morbidity and mortality. Influenza viruses replicate in the respiratory mucosa, primarily in respiratory epithelial cells producing large amounts of progeny virions, which then infect alveolar macrophages and resident dendritic cells.

Three types of influenza viruses classified as A, B and C infect humans [1,2]. While influenza A and B viruses (IAV and IBV, respectively) co-circulate annually during seasonal epidemics, influenza C viruses generally cause mild infections in humans. Of the three, only IAV have established animal reservoirs [2,3], although IBV have been isolated from some animal species [4,5]. Importantly, some swine and avian IAV strains can infect humans. Thus, seasonal IAV/IBV infections present a significant global health problem causing up to 500 000 excess deaths annually due to influenza infection [6,7]. IAVs are further distinguished into subtypes by their surface glycoproteins hemagglutinin (H) and neuraminidase (N) with 18 HA subtypes and 11 NA subtypes identified to date. On the other hand, IBVs are grouped into two antigenically and genetically distinct lineages, called B/Yamagata and B/Victoria [4,5]. Two IAV subtypes (H1N1 and H3N2), along with one or two IBV viruses from the two lineages co-circulate annually causing influenza epidemics. Under immune pressure, influenza viruses acquire mutations in their HA and NA proteins, which allow immune escape in a process known as antigenic drift. Novel IAVs emerge by reassortment, mostly in animal reservoirs, in a process known as antigenic shift. In case when they acquire the ability to spread from person-to-person, novel IAVs can cause a pandemic [2,3].

The first leukocytes that infiltrate the lungs following influenza virus infection are innate cells that include neutrophils and macrophages [8]. They are followed by influenza-specific B and T cells that mediate the adaptive immune response (Fig.1). While antibody-mediated immunity is mainly strain-specific and can prevent infection, T cells (especially CD8⁺ T cells) provide broadly cross-reactive cellular immunity against different influenza viruses and thus can ameliorate the severity of influenza disease [9–13]. Current vaccination strategies with inactivated influenza viruses induce strain-specific, B cell-based protective immunity directed against the viral hemagglutinin protein [14]. However, these vaccines offer little cross-protection against antigenic drift variants of seasonal IAVs and IBVs. Therefore, the vaccine formulation needs to be updated annually to cover antigenically drifted IAV and IBV strains. Notably, current seasonal influenza vaccines do not protect against new and unanticipated influenza viruses, such as pandemic strains.

T cells recognize epitopes on conserved internal viral proteins and thus they confer broader protection across IAV subtypes [9,15]. Ideally, vaccination strategies that induce broadly cross-protective T cell responses across antigenic drift variant influenza viruses, and potentially against newly emerging zoonotic IAVs, are required to limit the disease burden of influenza epidemics and pandemics. Here, we summarize the current knowledge on the innate and adaptive T cell responses to influenza viruses and discuss how these responses can be exploited by vaccines to prevent or ameliorate severe influenza infections.

Innate immunity as the first barrier against influenza viruses

The innate immune response involves powerful antiviral activity mediated by interferon (IFN), which limits early viral spread and subsequently shapes the adaptive immune response by their cytokine secretion [16]. Innate immunity to influenza viruses also involves several innate T cell subsets such as mucosal-associated invariant (MAIT), natural killer (NKT) and gd T cells, which have been a focus of recent studies.

Mucosal-associated invariant T (MAIT) cells contribute to immunity against influenza viruses

MAIT cells constitute up to 5% of the total T cell pool in humans and are characterized by the expression of a semi-invariant T cell receptor (TCR) Vα7.2 that recognizes bacterial-derived vitamin B2 derivatives presented by the MHC-like protein 1 (MR1) [17]. On activation, MAIT cells produce proinflammatory cytokines and upregulate cytotoxic granzymes and perforin [18]. Although MAIT cells are predominantly associated with control of bacterial infections, there is increasing evidence of their clinical importance in viral infections, autoimmune diseases, and cancers [19–26]. With respect to human IAV infection, van Wilgenburg et al. showed that the number of human MAIT cells in the peripheral blood of severely ill individuals was reduced by nearly half and they were highly activated, as shown by upregulation of granzyme B [26]. Furthermore, Loh et al. showed that patients who died from severe H7N9 infection had ~3–5-fold fewer MAIT cells in the blood, compared to patients who recovered from severe H7N9 infection or were healthy [25]. These studies suggest a loss and/or efflux of MAIT cells from the blood to the site of infection during severe influenza disease. Further, both studies showed that the MAIT cell activation in influenza infection was MR1-TCR independent and was mediated by IL-18 [25,26], mainly produced by CD14⁺ monocytes [25]. Reduction in MAIT cell numbers during acute influenza infection could impair protective anti-bacterial immunity increasing the risk of bacterial co-infection, which would enhance disease severity and mortality [27].

NKT cells modulate both innate and adaptive response mechanisms during influenza infection

Natural killer T cells are innate T lymphocytes restricted by the glycolipid α-galactosylceramide (αGC) presented on MHC class I-like CD1d molecules. Like conventional T cells, they express ab–TCRs as well as NK cell markers including CD3, CD56 (humans) or NK1.1 (mice). NKT cells constitute approximately 0.1%−0.2% of lymphocytes in human PBMCs and are broadly classified into 2 groups: type I invariant NKT (iNKT) cells, expressing the invariant Vα14-Jα18 TCRα chain paired with either Vβ2, Vβ7 or Vβ8 in mice and Vα24-Jα18/Vβ11 in humans, and type II diverse NKT cells, which display very diverse ab–TCR gene pairings [28,29]. In animals, NKT cells can protect against viral infections and cancer [28,30] by regulating innate and adaptive immune responses via rapid secretion of cytokines [31,32], upregulation of CD40, and activation of dendritic cells [33].

In a mouse model of influenza infection, intraperitoneal (i.p.) administration of aGC along with intranasal (i.n.) administration of an influenza virus (H1N1/H3N2 recombinant) activates iNKT cells (without affecting the adaptive T cell responses), consequently reducing body weight loss and viral lung titers [34]. Conversely, CD1d^−/− mice lacking the CD1d molecule (and therefore, lacking CD1d-stimulation of NKT cells) developed more severe disease following infection with a sublethal dose of influenza (PR8-H1N1) virus, as evidenced by lower body weight, higher lung viral titers, decreased IFNg production in the bronchoalveolar lavage (BAL), and impaired NK and T cell activity, as compared to wild type (WT) mice [35]. Similarly, Ja18^−/- mice (lacking NKT cells expressing Ja18) infected with the virulent influenza PR8-H1N1 virus displayed increased lung injury and mortality, associated with greater infiltration of Ly6C^hiLy6G^– inflammatory monocytes in lungs, compared to Ja18^−/− mice receiving aGC-boosted iNKT cells. Moreover, aGC-boosted iNKT cells reduced MCP-1 levels by lysing influenza-infected monocytes in a CD1d-dependent manner [36]. These findings were consistent with another study, in which a rapid reduction in body weight and 100% mortality by 11 days post-infection (d.p.i.) was observed in Ja18^−/− mice. The enhanced pathology in influenza-infected Ja18^−/− mice was associated with a ~4-fold decrease in D^bPA_224–233-specific CD8⁺ T cells in lungs at 4 and 7 d.p.i. and differential accumulation and maturation of CD103⁺ dendritic cells (DCs) [37]. Moreover, the absence of iNKT cells led to suppression of IAV responses due to increased expression of the arginase and nitric oxide system and this was mitigated by adoptive transfer of iNKT cells that restored antiviral immunity in a CD1d- and CD40-dependent manner [38]. Furthermore, the use of aGC as a vaccine adjuvant improved long-term survival of CD8⁺ cytotoxic T lymphocytes and provided heterologous protection in a mouse model of influenza infection [39]. Similarly, aGC analogs, KBC-007 and KBC-009, administered with inactivated influenza A/PR/8/34 virus boosted humoral and cellular immune responses in mucosal and systemic compartments post immunization and induced protective immune responses when challenged with PR8 virus [40]. Intramuscular administration of inactivated A/California/04/2009 pandemic H1N1 virus with aGC to piglets increased NKT-cell numbers, induced high titers of anti-HA mAbs, and boosted IAV-specific CD8⁺ T cells post-vaccination. Virus challenge in piglets vaccinated with aGC/inactivated virus resulted in reduced viral shedding in comparison to piglets vaccinated with whole inactive virus alone [41]. Intranasal immunization with inactivated influenza A/PR/8/34 virus with aGC in BALB/c mice induced IgG and IgA antibodies lasting up to 3 months and boosted cellular immunity in comparison to mice immunized with inactivated PR8 virus alone [42].

Collectively, these animal studies demonstrate the role played by NKT cells in modulating innate and adaptive immune responses during IAV infection. aGC as a vaccine adjuvant boosted NKT numbers and improved protection, further supporting the role of NKT cells in ensuring optimal influenza-specific T and B cell responses in the mucosal and systemic compartments. Despite the large number of animal studies investigating the role of NKT cells during influenza infection, there is a paucity of data supporting their protective role in human influenza, and this warrants further investigation.

Anti-viral potential of gd T cells during influenza virus infection

gd T cells (approximately 1%−10% of peripheral blood T cells in humans) are important components of both innate and adaptive immunity, displaying a vast array of effector functions, making them attractive targets for therapeutic vaccines. gd T cells are characterized by a diverse range of TCRs, with specific Vg-Vd pairings found across different anatomical sites. The antigen specificity of gd T cells remains largely unclear. While butyrophilin (BTN) protein family molecules can regulate immunosurveillance of gd T cells (reviewed in Ref. 43), phosphoantigens (pAg, recognized via a mechanism involving BTN3A1), human MutS homolog 2 and F1-ATPase along with apolipoprotein can activate Vg9Vd2 T cells, while CD1d or MICA-lipid complexes and endothelial protein C receptor can stimulate non-Vd2 T cells. gd T cells can directly kill virally-infected, cancerous or stressed cells by secreting perforin, granzyme B, and granulysin, and can promote inflammation and wound healing, and assist antigen presenting cells in generating optimal, antigen-specific T cell responses (reviewed in Refs. 43 and 44). Murine studies of influenza infection showed that 15%−30% of the lymphocytes in the BAL between 10 and 15 d.p.i. were gd T cells, compared to very few at 5−7 d.p.i. This suggests that gd T cells may play a pivotal role in recovery from influenza infection [45]. Moreover, in a secondary influenza infection model, mice that were challenged with an H1N1 virus a month following H3N2 virus infection, gd T cells isolated from the BAL acquired an NK1.1 phenotype and displayed potent cytotoxic functions, as shown by killing of H3N2 virus-infected target cells in the presence of PHA [46]. Moreover, gd T cell hybridomas, derived from TCRβ^−/− mouse, cleared IAV and IBV infections efficiently via Hsp60 upregulation in infected cells and not by any viral proteins [47]. In humans, ex-vivo gd T cells rapidly produced IFNg and upregulated CD25 and CD69 following IAV infection. The activation of gd T cells is dependent on the mevalonate pathway, as shown by a potent reduction in antiviral responses on addition of mevastatin, a mevalonate pathway inhibitor [48]. Furthermore, in vitro studies with human peripheral blood mononuclear cells (PBMC) showed that pAg-activated gd T cells were capable of killing macrophages and lung alveolar epithelial cells infected with both human (H1N1) and avian (H5N1) influenza viruses [49–52]. Moreover, pAg isopentenyl pyrophosphate (IPP)-expanded human Vg9Vd2 T cells displayed cytolytic activity (as measured by % target cell death), increased production of IFNg against pandemic H1N1 virus-infected monocyte-derived macrophages, and expressed the inflammatory chemokine receptors CCR1, CCR5, and CXCR5 [50]. pAg-reactive Vg9Vd2 T subsets harbour a semi-invariant Vg9 pairing with diverse Vd2 TCRs [53]. Vg9Vd2 T cells can get stimulated prior to exposure with environmental antigens, as exemplified by fetal blood-derived Vg9Vd2 T cells and their potent activation following pAg exposure [54]. Overall, these human studies implicate that the pAg-reactive Vg9Vd2 T subsets harbour innate-like activity toward influenza infection. These findings highlight the anti-viral potential of gd T cells during influenza infection and suggest that enhancement of gd T cell responses might be beneficial in protection against influenza viruses.

Adaptive T cells are crucial for recovery and long-term protection from influenza infection

Following the initial wave of innate immunity, the adaptive immune system becomes activated and plays an important role in viral clearance, host recovery, and establishment of immunological memory. The adaptive immune system consists of humoral (virus-specific antibodies) and cellular (CD4⁺ and CD8⁺ T cells) components. Here, we highlight the cellular aspects of influenza-specific adaptive immunity and discuss how the current vaccine strategies target memory T cells.

CD8⁺ T cells confer cross-protection across different influenza subtypes

Cytotoxic CD8⁺ T cells promote efficient elimination of the virus and subsequent host recovery, via the production of anti-viral cytokines and direct killing of virus-infected cells [55,56]. Influenza-specific CD8⁺ T cells play a key role in broadly cross-reactive immunity to influenza viruses because they recognize 8–11 amino acids long peptide fragments derived from highly conserved internal viral proteins, presented in the context of MHC class-I molecules. CD8⁺ T cell-mediated cross-protection against different influenza subtypes has been well established in animal studies, including delayed virus clearance in mice lacking CD8⁺ T cells [55], generation of influenza virus escape mutants in influenza-infected transgenic mice expressing influenza-specific CD8⁺ TCR [57], or even more strikingly, in wild-type MHC-I-matched (but not MHC-I mismatched) mice during acute influenza infection [58]. When mice that have recovered from a primary influenza infection are re-infected with a heterologous IAV (a strain with different surface glycoproteins but the same internal proteins) that is resistant to neutralization by antibodies induced by the primary infection, influenza-specific CD8⁺ T cells are responsible for accelerated viral clearance and reduced disease burden [13,59–61]. Studies in guinea pigs also demonstrated that priming with the 2009 pandemic H1N1 virus confers protection against the 2013 avian H7N9 IAV [62]. Recently, influenza-specific tissue-resident memory (T_RM) CD8⁺ T cells in the upper respiratory tract of mice were shown to be important in mediating protection from pulmonary influenza virus infection [63]. The presence of protective CD8⁺ T_RM populations in the nasal tissue implies that these cytotoxic effectors are in place to contribute to the early control of infection. Indeed, in mouse studies the establishment of CD8⁺ T_RM cells in the upper respiratory tract reduced viral replication in the nose as early as day 3 and day 5 after intranasal challenge, in a CD8⁺ T cell-dependent manner [63].

CD8⁺ T cells are also associated with protection from disease during human IAV infection. In a seminal paper by Andrew McMichael and colleagues [9], healthy volunteers were experimentally infected with the influenza A/Munich/1/79 (H1N1) virus. Individuals with demonstrable influenza-specific CD8⁺ T cell responses cleared the virus more efficiently, irrespective of the presence of pre-existing antibodies [9]. Subsequently, a retrospective report showed that individuals previously infected with H1N1 IAV in the 1950s were less susceptible to the 1957 H2N2 pandemic [10], further supporting the idea of heterosubtypic immunity that is presumed to be conferred by cellular immunity because antibodies elicited by H1N1 viruses do not protect against H2N2 IAV infection. The advent of the 2009 H1N1 pandemic provided further opportunities to determine the role of CD8⁺ T cells in protection against IAV infection. Indeed, in a household cohort study in the UK, pre-existing IFNg-producing CD8⁺ T cell responses were associated with lower disease severity [12]. Furthermore, the Flu Watch study, recruiting influenza-infected patients in England between 2006 and 2010, showed a positive correlation between pre-existing IAV-specific CD8⁺ T cells and less symptomatic, PCR-confirmed IAV infection during 3 seasonal epidemics and the 2009 H1N1 pandemic [64]. These studies provide compelling evidence that cytotoxic CD8⁺ T cells can confer some level of protection during infection with human influenza viruses. Recently, our group has extended these observations to infection with the avian H7N9 IAV that emerged in China in 2013 in a study of longitudinal samples from hospitalized patients during the first wave of H7N9 infections in China [13]. Patients with most rapid recovery (discharged within 2–3 weeks after hospitalization) had early and robust H7N9-specific CD8⁺ T cell responses, followed later by neutralizing antibody responses. Conversely, patients who succumbed to infection had delayed and/or minimal cytotoxic CD8⁺ T cell responses. Thus, recovery from severe H7N9 infection was associated with robust early CD8⁺ T cell responses.

Longevity and antigen-specificity of influenza-specific CD8⁺ T cells

The ability of CD8⁺ T cells to provide cross-strain protection against distinct influenza viruses makes them promising and attractive vaccine targets. However, one needs to also consider the longevity and breadth of the CD8⁺ T cell response. Indeed, influenza-specific CD8⁺ T cells were stable in the peripheral blood of healthy adults over a period of 13 years, showing that memory CD8⁺ T cells are long-lived [65]. The breadth of CD8⁺ T cell responses results from the high degree of conservation of immunogenic epitopes. Recent human studies showed that CD8⁺ T cells can confer cross-reactivity across multiple IAVs [66–68], including H7N9 [69,70] and H5N1 [71,72] viruses, as well as the two lineages of IBV [73]. Our group has recently identified a set of universal and immunodominant CD8⁺ T cell epitopes (HLA-A*02:01/M1_58-66, HLA-A*03:01/NP_265-273, HLA-B*08:01/NP_225-233, HLA-B*18:01/NP_219-226, HLA-B*57:01/NP_199-207), which are conserved across 100% of known human IAVs (from the 1918 pandemic to the 2013 H7N9 strain) [69]. These epitopes encompass various HLA molecules that are differentially prevalent across ethnicities, being particularly rare in the Alaskan and Australian Indigenous populations [74]. It is pertinent to note that cross-reactivity may occur even when new variant epitopes emerge, depending on the amino acid substitution and/or position of the mutation and its impact on TCR recognition of the peptide/MHCI complex as well the individual’s TCR repertoire of the cognate CD8⁺ T cell population. For instance, mutations in anchor residues can allow for escape variants, as it was the case with HLA-B*27:05/NP_383-391, whereby mutations in position 2 (R384G) abrogated binding of the peptide to the MHCI molecule [75]. Furthermore, for the HLA-B*07:02 or HLA-B*35:01 presented NP_418-426 peptide, while the anchor residues are maintained, cross-reactivity depends on the position of the mutation in the peptide and the nature of the substitution [68,76]. Conversely, in the context of HLA-A*02:01/M1_58-66, for which sporadic mutations have arisen over the last century, the public ab–TCR clonotype (TRBV19/TRAV27) is able to robustly cross-react with all the variants through induced-fit molecular mimicry, although in some instances with lower affinity [67]. These studies highlight the importance of the TCR repertoire in determining the breadth and protective capacity of the cytotoxic CD8⁺ T cell responses.

While CD8⁺ T cells are crucial in conferring protection, excessive CD8⁺ T cell responses can lead to immunopathology, as recently reviewed by Duan and Thomas [77]. However, the exact molecular and cellular pathways that lead to regulated and effective CD8⁺ T cell responses versus exuberant and pathogenic responses are not entirely clear [78]. Thus, future research is needed to understand how protective memory can be established without immunopathology. It is also important to note that CD8⁺ T cells do not provide sterilizing immunity against infection but promote rapid host recovery and thereby provide protection from disease symptoms and death. This is why CD8⁺ T cells play a pivotal role in reducing disease severity and preventing fatal infections especially during an outbreak of a novel antigenically-shifted influenza strain, to which there is no pre-existing neutralizing immunity in the general population.

Thus, it is evident that CD8⁺ T cells can confer broad protective immunity to antigenically drifted and shifted influenza viruses. The HLA genotype and the TCR repertoire are two key factors that need to be considered in the design of T cell-targeted vaccine strategies to provide wide population coverage, while limiting immune escape.

New insights into the role of CD4⁺ T cells during influenza virus infection

CD4⁺ T cells (T helper cells, Th cells) commonly recognize viral peptide fragments that are 12–16 amino acids in length, presented on major histocompatibility complex class-II (MHC-II) molecules. They play an important role in promoting optimal CD8⁺ T cell and B cell responses during influenza infection. Following activation in the draining lymph nodes during influenza infection, naïve CD4⁺ T cells differentiate into effector cells capable of producing cytokines such as IFNg and IL-2, and migrate to the lung. This is accompanied by the loss of the homing receptor CD62L, downregulation of CCR7 and upregulation of CD44 and CD49d, with transient expression of CCR5 [79]. CD4⁺ T cells, especially T follicular helper cells (Tfh) [80,81], help B cells to generate antibodies by providing co-stimulatory signals. They are important for the priming and maintenance of CD8⁺ T cell responses to generate efficient memory [82,83].

Circulating T follicular helper cells (cTfh; CXR5⁺ PD1⁺ CD4⁺ T cells) have recently been identified during influenza infection and vaccination [84] as a prominent CD4⁺ T cell population in peripheral blood on day 7 after vaccination [85]. These cTfh cells express the Th1 phenotype (CXCR3⁺CCR6⁻) and upregulate markers of activation ICOS and PD-1. Importantly, cTfh cells can provide help in the form of cytokines and co-stimulation signals to memory B cells for development into antibody-secreting cells (ASCs) in vitro [85]. As a result, the presence of cTfh cells correlates with the magnitude and quality of the antibody response and the development of ASCs following vaccination with trivalent inactivated vaccines (TIV) [85,86]. The ontogeny and specificity of cTfh cells and their relationship to memory Tfh reservoirs in lymphoid tissues remain unclear. Recently, clonotypic analysis of cTfh cells revealed the recruitment of recurrent clonotypes following vaccination in multiple years, strongly indicating that the cTfh population that appears following vaccination is memory-derived [87]. Further understanding of the specificity and function of circulating and non-circulating Tfh cells in humans and the role of these cells in influenza infection may inform the development of improved vaccines.

CD4⁺ T cells assist CD8⁺ T cells in becoming activated by licensing Ag-presenting DCs via the CD154 (CD40L)/CD40 signaling pathway [88,89]. However, CD4⁺ T cell-mediated help for primary activation of naïve CD8⁺ T cells appears to be pathogen-dependent, as it was redundant during influenza virus or LCMV infections [90]. Efficient primary CD8⁺ T cell responses could be established in the absence of CD4⁺ T cell help, as shown in mice depleted of CD4⁺ T cells or in MHC-II knock-out mice [91,92]. Nevertheless, co-stimulatory signaling via CD28 is needed for primary CD8⁺ T cell activation [93] and is required to sustain CD8⁺ T cell responses after activation during the effector phase [94]. Furthermore, virus-specific CD4⁺ T cells have a beneficial impact on primary and secondary responses for themselves, as CD4⁺ T cell licensing of dendritic cells results in enhanced proliferation and differentiation of the cells in influenza infected mice [95].

Regardless of whether CD4⁺ T cell help is required for primary influenza-specific CD8⁺ T cell responses, they are beneficial for generation and recall of CD8⁺ T cell memory in terms of numbers of antigen-specific CD8⁺ T cells after the secondary influenza virus infection [82,96]. Furthermore, CD4⁺ T cell help is needed to localize tissue-resident memory CD8⁺ T cells to epithelial cells in the lung, as influenza infection in mice lacking help led to less expression of the integrin CD103 and a dislocation of T_RMs at 45 d.p.i. [97]. CD4⁺ T cell help during influenza priming of CD8⁺ T cells led to greater production of granzyme B (GzmB) in the T_RMs after secondary challenge [97].

In addition to their helper function for B cell antibody class switching [98] and licensing of DCs for CD8⁺ T cell activation, CD4⁺ T cells can acquire cytotoxic functions [99,100]. Notably, acquisition of cytotoxic functions after influenza virus infection in mice was strongly tissue-dependent. Therefore, in adoptive transfer studies, cytotoxic CD4⁺ T cells expressing GzmB and perforin could be found preferentially in mucosal tissue such as the lungs [101] of influenza (PR8-Ova) infected mice but less often in draining mediastinal lymph nodes and spleen at 7 d.p.i. [101]. Furthermore, the route of administration of the virus appears to be important: intranasal infection of mice with PR8 IAV resulted in recruitment of the cytolytic CD4⁺ T cells in the lungs, while intramuscular administration led to diminished lung cytolytic CD4⁺ T cells and sterilizing immunity [102]. Thus, the route of the administration is an important consideration for future vaccine approaches designed to elicit a protective immune barrier at the site of infection.

Ex vivo analysis of influenza-specific CD4⁺ T cells has been recently facilitated by the development of HLA class II tetramers, generally characterized by lower TCR binding avidity to peptide/MHCII compared to tetramer-specific CD8⁺ T cells. When cross-reactivity to four strains, including the pandemic H1N1 virus was analyzed in humans, antigen-specific CD4⁺ T cells were detected against several influenza viral proteins, including the hemagglutinin (H1 and H3), neuraminidase (N1 and N2), nucleoprotein (NP), and Matrix 1 (M1) [15,72,103]. Cross-reactivity of influenza-specific memory CD4⁺ T cells was also shown using overlapping peptides derived from influenza A/Vietnam/CL26/2005 (H5N1) and influenza A/New York/232/2004 (H3N2) in healthy volunteers from the UK and Vietnam. The viral M1 and NP proteins were the immunodominant targets of this CD4⁺ T cell cross-recognition [72]. Further research is needed to understand how CD4⁺ T cell help leads to the generation of efficient CD4⁺ and CD8⁺ T cell memory.

Vaccine strategies for induction of broadly cross-reactive adaptive T cell responses

Currently, inactivated influenza vaccines (IIV) remain the predominant strategy for inducing protective immunity for upcoming epidemics. The IIV, either in a trivalent (H1N1, H3N2, and one IBV from either lineage) or quadrivalent (H1N1, H3N2, and two IBV strains, one from each lineage) formulation, is prepared annually based on recommendations by the WHO. Although IIV induces a B cell and antibody response in primed individuals, it is a weak inducer of T cell immunity, with the exception of CD4⁺ T cells.

Promising approaches for targeting CD4⁺ T cells by vaccination include the use of oil-in-water adjuvants such as MF59® adjuvant [104]. Another approach is the live-attenuated influenza virus vaccine (LAIV), which in contrast to injectable IIV, can generate long-term and virus-specific lung tissue-resident memory CD4⁺ and CD8⁺ T cells in mice [105]. However, human studies on LAIV show minimal increases in CD8⁺ T cell responses after vaccination, as measured by IFNg production [106,107]. Nevertheless, it remains unclear whether the poor results observed in these human LAIV trials are due to the fact that CD8⁺ T cells are being assessed within the peripheral blood, while memory T cell responses generated after LAIV in mice are established at the site of infection [108].

Given the prominent role of T cells in inducing protective and universal immunity, novel T cell-mediated vaccination strategies are required to complement the induction of antibodies and memory B cells. One novel vaccine formulation is the modified vaccinia virus Ankara (MVA) expressing the influenza viral proteins NP and M1. The MVA-NP+ M1 vaccine can induce increases in the number of influenza-specific IFNg-secreting CD8⁺ T cells in adults [109]. Additionally, the number of tetramer⁺ HLA-A*02:01/M1_58-66–specific CD8⁺ T cells was shown to increase and upregulate the effector molecules granzyme A and perforin [110]. Importantly, compared to unvaccinated controls, the MVA-NP+ M1 vaccine provided modest protection from clinical symptoms upon intranasal challenge with influenza A/Wisconsin/67/2005 (H3N2) [111]. Other vaccine formulations have incorporated influenza peptides into their formulations, such as Flu-v [112,113] and FP-0.01 [113], or the Multimeric-001 vaccine, which consists of a protein encompassing 9 highly conserved peptides from the M1, NP, and HA proteins [114]. Individuals vaccinated with Flu-v showed a modest increase in IFNg⁺ T cell responses following in vitro stimulation with peptides [112,115], and these responses correlated with reduced viral titers following challenge with A/Wisconsin/67/2005 (H3N2). However, there was no correlation between IFNg⁺ T cell responses and clinical symptoms. Consistently, there were no significant differences in viral shedding or symptom score between vaccinated and unvaccinated volunteers following experimental challenge [115]. The FP-0.01 and Multimeric-001 vaccines only show small increases in T cell responses following restimulation of PBMCs from vaccinated individuals and have yet to be tested for the protective efficacy against experimental IAV challenge [113,114]. Table 1 shows an overview of recent trials that target innate and adaptive T cells for influenza vaccination [39–42,104,105,107,109,111–116].

Overall, there are currently no vaccines that induce robust and protective CD8⁺ T cell responses against influenza viruses. Novel vaccination strategies that can establish broadly cross-reactive and long-lived memory CD8⁺ T cells are needed in order to complement antibody responses in the prevention of seasonal influenza and newly emerging influenza viruses.

Concluding remarks

Published studies show clear evidence that both the adaptive (T and B cells) and innate (MAIT cells, gd T cells, and NKT cells) immune responses play important roles in controlling influenza virus infection. Thus, vaccination strategies that link innate and adaptive cellular immune responses would be attractive. Currently licensed vaccine strategies do not induce robust and protective T cell responses. The development of a vaccine that induces cross-protective cellular immunity against different influenza strains and subtypes, including IAVs with pandemic potential, would be an excellent achievement.

References

[1] Bouvier NM, Palese  P. The biology of influenza viruses. Vaccine 2008; 26(Suppl 4): D49–D53

[2] Paules C, Subbarao  K. Influenza. Lancet 2017; 390(10095): 697–708 

[3] Shaw M, Palese  P. Orthomyxoviridae: the viruses and their replication. In: Fields Virology. 6th edition. Edited by Knipe D, Howley P. Lippincott Williams & Wilkins, 2013:1691–1740

[4] Koutsakos M, Nguyen  TH, Barclay WS,  Kedzierska K. Knowns and unknowns of influenza B viruses. Future Microbiol 2016; 11(1): 119–135

[5] van de Sandt CE,  Bodewes R,  Rimmelzwaan GF,  de Vries RD. Influenza B viruses: not to be discounted. Future Microbiol 2015; 10(9): 1447–1465

[6] WHO. A manual for estimating disease burden associated with seasonal influenza. 2015

[7] WHO. Fact sheet, No. 211 (2009).  2010

[8] La Gruta NL, Kedzierska  K, Stambas J,  Doherty PC. A question of self-preservation: immunopathology in influenza virus infection. Immunol Cell Biol 2007; 85(2): 85–92

[9] McMichael AJ, Gotch  FM, Noble GR,  Beare PA. Cytotoxic T-cell immunity to influenza. N Engl J Med 1983; 309(1): 13–17

[10] Epstein SL. Prior H1N1 influenza infection and susceptibility of Cleveland Family Study participants during the H2N2 pandemic of 1957: an experiment of nature. J Infect Dis 2006; 193(1): 49–53

[11] Wilkinson TM, Li  CK, Chui CS,  Huang AK,  Perkins M,  Liebner JC,  Lambkin-Williams R,  Gilbert A,  Oxford J,  Nicholas B,  Staples KJ,  Dong T, Douek  DC, McMichael AJ,  Xu XN. Preexisting influenza-specific CD4⁺ T cells correlate with disease protection against influenza challenge in humans. Nat Med 2012; 18(2): 274–280

[12] Sridhar S, Begom  S, Bermingham A,  Hoschler K,  Adamson W,  Carman W,  Bean T, Barclay  W, Deeks JJ,  Lalvani A. Cellular immune correlates of protection against symptomatic pandemic influenza. Nat Med 2013; 19(10): 1305–1312

[13] Wang Z, Wan  Y, Qiu C,  Quiñones-Parra S,  Zhu Z, Loh  L, Tian D,  Ren Y, Hu  Y, Zhang X,  Thomas PG,  Inouye M,  Doherty PC,  Kedzierska K,  Xu J. Recovery from severe H7N9 disease is associated with diverse response mechanisms dominated by CD8⁺ T cells. Nat Commun 2015; 6: 6833

[14] Baumgarth N, Herman  OC, Jager GC,  Brown LE,  Herzenberg LA,  Chen J. B-1 and B-2 cell-derived immunoglobulin M antibodies are nonredundant components of the protective response to influenza virus infection. J Exp Med 2000; 192(2): 271–280

[15] Assarsson E, Bui  HH, Sidney J,  Zhang Q,  Glenn J,  Oseroff C,  Mbawuike IN,  Alexander J,  Newman MJ,  Grey H, Sette  A. Immunomic analysis of the repertoire of T-cell specificities for influenza A virus in humans. J Virol 2008; 82(24): 12241–12251

[16] Randall RE, Goodbourn  S. Interferons and viruses: an interplay between induction, signalling, antiviral responses and virus countermeasures. J Gen Virol 2008; 89(1): 1–47

[17] Treiner E, Lantz  O. CD1d- and MR1-restricted invariant T cells: of mice and men. Curr Opin Immunol 2006; 18(5): 519–526

[18] Howson LJ, Salio  M, Cerundolo V. MR1-restricted mucosal-associated invariant T cells and their activation during infectious diseases. Front Immunol 2015; 6: 303

[19] Kawachi I, Maldonado  J, Strader C,  Gilfillan S. MR1-restricted V alpha 19i mucosal-associated invariant T cells are innate T cells in the gut lamina propria that provide a rapid and diverse cytokine response. J Immunol 2006; 176(3): 1618–1627

[20] Napier RJ, Adams  EJ, Gold MC,  Lewinsohn DM. The role of mucosal associated invariant T cells in antimicrobial immunity. Front Immunol 2015; 6: 344

[21] Sakala IG, Kjer-Nielsen  L, Eickhoff CS,  Wang X, Blazevic  A, Liu L,  Fairlie DP,  Rossjohn J,  McCluskey J,  Fremont DH,  Hansen TH,  Hoft DF. Functional heterogeneity and antimycobacterial effects of mouse mucosal-associated invariant T Cells specific for riboflavin metabolites. J Immunol 2015; 195(2): 587–601

[22] Bianchini E, De Biasi  S, Simone AM,  Ferraro D,  Sola P, Cossarizza  A, Pinti M. Invariant natural killer T cells and mucosal-associated invariant T cells in multiple sclerosis. Immunol Lett 2017; 183: 1–7

[23] Gracey E, Qaiyum  Z, Almaghlouth I,  Lawson D,  Karki S,  Avvaru N,  Zhang Z,  Yao Y, Ranganathan  V, Baglaenko Y,  Inman RD. IL-7 primes IL-17 in mucosal-associated invariant T (MAIT) cells, which contribute to the Th17-axis in ankylosing spondylitis. Ann Rheum Dis 2016; 75(12): 2124–2132

[24] Wang JJ, Macardle  C, Weedon H,  Beroukas D,  Banovic T. Mucosal-associated invariant T cells are reduced and functionally immature in the peripheral blood of primary Sjögren’s syndrome patients. Eur J Immunol 2016; 46(10): 2444–2453

[25] Loh L, Wang  Z, Sant S,  Koutsakos M,  Jegaskanda S,  Corbett AJ,  Liu L, Fairlie  DP, Crowe J,  Rossjohn J,  Xu J, Doherty  PC, McCluskey J,  Kedzierska K. Human mucosal-associated invariant T cells contribute to antiviral influenza immunity via IL-18-dependent activation. Proc Natl Acad Sci USA 2016; 113(36): 10133–10138

[26] van Wilgenburg B,  Scherwitzl I,  Hutchinson EC,  Leng T, Kurioka  A, Kulicke C,  de Lara C,  Cole S, Vasanawathana  S, Limpitikul W,  Malasit P,  Young D,  Denney L,  Moore MD,  Fabris P,  Giordani MT,  Oo YH, Laidlaw  SM, Dustin LB,  Ho LP, Thompson  FM, Ramamurthy N,  Mongkolsapaya J,  Willberg CB,  Screaton GR,  Klenerman P. MAIT cells are activated during human viral infections. Nat Commun 2016; 7: 11653

[27] McCullers JA. The co-pathogenesis of influenza viruses with bacteria in the lung. Nat Rev Microbiol 2014; 12(4): 252–262

[28] Godfrey DI, Uldrich  AP, McCluskey J,  Rossjohn J,  Moody DB. The burgeoning family of unconventional T cells. Nat Immunol 2015; 16(11): 1114–1123

[29] Rossjohn J, Pellicci  DG, Patel O,  Gapin L,  Godfrey DI. Recognition of CD1d-restricted antigens by natural killer T cells. Nat Rev Immunol 2012; 12(12): 845–857

[30] Tessmer MS, Fatima  A, Paget C,  Trottein F,  Brossay L. NKT cell immune responses to viral infection. Expert Opin Ther Targets 2009; 13(2): 153–162

[31] Crowe NY, Uldrich  AP, Kyparissoudis K,  Hammond KJL,  Hayakawa Y,  Sidobre S,  Keating R,  Kronenberg M,  Smyth MJ,  Godfrey DI. Glycolipid antigen drives rapid expansion and sustained cytokine production by NK T cells. J Immunol 2003; 171(8): 4020–4027

[32] Coquet JM, Chakravarti  S, Kyparissoudis K,  McNab FW,  Pitt LA,  McKenzie BS,  Berzins SP,  Smyth MJ,  Godfrey DI. Diverse cytokine production by NKT cell subsets and identification of an IL-17-producing CD4-NK1.1-NKT cell population. Proc Natl Acad Sci USA 2008; 105(32): 11287–11292

[33] Vincent MS, Leslie  DS, Gumperz JE,  Xiong X,  Grant EP,  Brenner MB. CD1-dependent dendritic cell instruction. Nat Immunol 2002; 3(12): 1163–1168

[34] Ho LP, Denney  L, Luhn K,  Teoh D, Clelland  C, McMichael AJ. Activation of invariant NKT cells enhances the innate immune response and improves the disease course in influenza A virus infection. Eur J Immunol 2008; 38(7):1913–1922. 

[35] Ishikawa H, Tanaka  K, Kutsukake E,  Fukui T,  Sasaki H,  Hata A, Noda  S, Matsumoto T. IFN-g production downstream of NKT cell activation in mice infected with influenza virus enhances the cytolytic activities of both NK cells and viral antigen-specific CD8⁺ T cells. Virology 2010; 407(2): 325–332

[36] Kok WL, Denney  L, Benam K,  Cole S, Clelland  C, McMichael AJ,  Ho LP. Pivotal advance: invariant NKT cells reduce accumulation of inflammatory monocytes in the lungs and decrease immune-pathology during severe influenza A virus infection. J Leukoc Biol 2012; 91(3): 357–368

[37] Paget C, Ivanov  S, Fontaine J,  Blanc F,  Pichavant M,  Renneson J,  Bialecki E,  Pothlichet J,  Vendeville C,  Barba-Spaeth G,  Huerre MR,  Faveeuw C,  Si-Tahar M,  Trottein F. Potential role of invariant NKT cells in the control of pulmonary inflammation and CD8⁺ T cell response during acute influenza A virus H3N2 pneumonia. J Immunol 2011; 186(10): 5590–5602

[38] De Santo C, Salio  M, Masri SH,  Lee LYH,  Dong T, Speak  AO, Porubsky S,  Booth S,  Veerapen N,  Besra GS,  Gröne HJ,  Platt FM,  Zambon M,  Cerundolo V. Invariant NKT cells reduce the immunosuppressive activity of influenza A virus-induced myeloid-derived suppressor cells in mice and humans. J Clin Invest 2008; 118(12): 4036–4048

[39] Guillonneau C, Mintern  JD, Hubert FX,  Hurt AC,  Besra GS,  Porcelli S,  Barr IG,  Doherty PC,  Godfrey DI,  Turner SJ. Combined NKT cell activation and influenza virus vaccination boosts memory CTL generation and protective immunity. Proc Natl Acad Sci USA 2009; 106(9): 3330–3335

[40] Lee YS, Lee  KA, Lee JY,  Kang MH,  Song YC,  Baek DJ,  Kim S, Kang  CY. An α-GalCer analogue with branched acyl chain enhances protective immune responses in a nasal influenza vaccine. Vaccine 2011; 29(3): 417–425

[41] Artiaga BL, Yang  G, Hackmann TJ,  Liu Q, Richt  JA, Salek-Ardakani S,  Castleman WL,  Lednicky JA,  Driver JP. α-Galactosylceramide protects swine against influenza infection when administered as a vaccine adjuvant. Sci Rep 2016; 6(1): 23593

[42] Youn HJ, Ko  SY, Lee KA,  Ko HJ, Lee  YS, Fujihashi K,  Boyaka PN,  Kim SH, Horimoto  T, Kweon MN,  Kang CY. A single intranasal immunization with inactivated influenza virus and alpha-galactosylceramide induces long-term protective immunity without redirecting antigen to the central nervous system. Vaccine 2007; 25(28): 5189–5198

[43] Kabelitz D, Lettau  M, Janssen O. Immunosurveillance by human gammadelta T lymphocytes: the emerging role of butyrophilins. F1000Res 2017; 6. pii: F1000 Faculty Rev-782

[44] Bonneville M, Scotet  E. Human Vgamma9Vdelta2 T cells: promising new leads for immunotherapy of infections and tumors. Curr Opin Immunol 2006; 18(5): 539–546

[45] Carding SR, Allan  W, Kyes S,  Hayday A,  Bottomly K,  Doherty PC. Late dominance of the inflammatory process in murine influenza by gamma/delta+ T cells. J Exp Med 1990; 172(4): 1225–1231

[46] Eichelberger M, Doherty  PC. Gamma delta T cells from influenza-infected mice develop a natural killer cell phenotype following culture. Cell Immunol 1994; 159(1): 94–102

[47] Wallace M, Malkovsky  M, Carding SR. Gamma/delta T lymphocytes in viral infections. J Leukoc Biol 1995; 58(3): 277–283

[48] Jameson JM, Cruz  J, Costanzo A,  Terajima M,  Ennis FA. A role for the mevalonate pathway in the induction of subtype cross-reactive immunity to influenza A virus by human gammadelta T lymphocytes. Cell Immunol 2010; 264(1): 71–77

[49] Qin G, Mao  H, Zheng J,  Sia SF, Liu  Y, Chan PL,  Lam KT, Peiris  JS, Lau YL,  Tu W. Phosphoantigen-expanded human gammadelta T cells display potent cytotoxicity against monocyte-derived macrophages infected with human and avian influenza viruses. J Infect Dis 2009; 200(6): 858–865

[50] Qin G, Liu  Y, Zheng J,  Ng IH, Xiang  Z, Lam KT,  Mao H, Li  H, Peiris JS,  Lau YL, Tu  W. Type 1 responses of human Vg9Vd2 T cells to influenza A viruses. J Virol 2011; 85(19): 10109–10116 

[51] Tu W, Zheng  J, Liu Y,  Sia SF, Liu  M, Qin G,  Ng IH, Xiang  Z, Lam KT,  Peiris JS,  Lau YL. The aminobisphosphonate pamidronate controls influenza pathogenesis by expanding a gammadelta T cell population in humanized mice. J Exp Med 2011; 208(7): 1511–1522

[52] Li H, Xiang  Z, Feng T,  Li J, Liu  Y, Fan Y,  Lu Q, Yin  Z, Yu M,  Shen C, Tu  W. Human Vg9Vd2-T cells efficiently kill influenza virus-infected lung alveolar epithelial cells. Cell Mol Immunol 2013; 10(2): 159–164

[53] Davey MS, Willcox  CR, Joyce SP,  Ladell K,  Kasatskaya SA,  McLaren JE,  Hunter S,  Salim M,  Mohammed F,  Price DA,  Chudakov DM,  Willcox BE. Clonal selection in the human Vd1 T cell repertoire indicates gd TCR-dependent adaptive immune surveillance. Nat Commun 2017; 8: 14760

[54] Dimova T, Brouwer  M, Gosselin F,  Tassignon J,  Leo O, Donner  C, Marchant A,  Vermijlen D. Effector Vgamma9Vdelta2 T cells dominate the human fetal gammadelta T-cell repertoire. Proc Natl Acad Sci USA 2015; 112(6):E556–65

[55] Bender BS, Croghan  T, Zhang L,  Small PAJ Jr. Transgenic mice lacking class I major histocompatibility complex-restricted T cells have delayed viral clearance and increased mortality after influenza virus challenge. J Exp Med 1992; 175(4): 1143–1145

[56] Doherty PC, Topham  DJ, Tripp RA. Establishment and persistence of virus-specific CD4⁺ and CD8⁺ T cell memory. Immunol Rev 1996; 150(1): 23–44

[57] Price GE, Ou  R, Jiang H,  Huang L,  Moskophidis D. Viral escape by selection of cytotoxic T cell-resistant variants in influenza A virus pneumonia. J Exp Med 2000; 191(11): 1853–1867

[58] Valkenburg SA, Quiñones-Parra  S, Gras S,  Komadina N,  McVernon J,  Wang Z, Halim  H, Iannello P,  Cole C, Laurie  K, Kelso A,  Rossjohn J,  Doherty PC,  Turner SJ,  Kedzierska K. Acute emergence and reversion of influenza A virus quasispecies within CD8⁺ T cell antigenic peptides. Nat Commun 2013; 4: 2663

[59] Duan S, Meliopoulos  VA, McClaren JL,  Guo XZ, Sanders  CJ, Smallwood HS,  Webby RJ,  Schultz-Cherry SL,  Doherty PC,  Thomas PG. Diverse heterologous primary infections radically alter immunodominance hierarchies and clinical outcomes following H7N9 influenza challenge in mice. PLoS Pathog 2015; 11(2): e1004642

[60] Valkenburg SA, Gras  S, Guillonneau C,  La Gruta NL,  Thomas PG,  Purcell AW,  Rossjohn J,  Doherty PC,  Turner SJ,  Kedzierska K. Protective efficacy of cross-reactive CD8⁺ T cells recognising mutant viral epitopes depends on peptide-MHC-I structural interactions and T cell activation threshold. PLoS Pathog 2010; 6(8): e1001039

[61] Valkenburg SA, Venturi  V, Dang TH,  Bird NL,  Doherty PC,  Turner SJ,  Davenport MP,  Kedzierska K. Early priming minimizes the age-related immune compromise of CD8⁺ T cell diversity and function. PLoS Pathog 2012; 8(2): e1002544

[62] Wiersma LC, Vogelzang-van Trierum  SE, Kreijtz JH,  van Amerongen G,  van Run P,  Ladwig M,  Banneke S,  Schaefer H,  Fouchier RA,  Kuiken T,  Osterhaus AD,  Rimmelzwaan GF. Heterosubtypic immunity to H7N9 influenza virus in isogenic guinea pigs after infection with pandemic H1N1 virus. Vaccine 2015; 33(49): 6977–6982

[63] Pizzolla A, Wang  Z, Groom JR,  Kedzierska K,  Brooks AG,  Reading PC,  Wakim LM. Nasal-associated lymphoid tissues (NALTs) support the recall but not priming of influenza virus-specific cytotoxic T cells. Proc Natl Acad Sci USA 2017; 114(20): 5225–5230

[64] Hayward AC, Wang  L, Goonetilleke N,  Fragaszy EB,  Bermingham A,  Copas A,  Dukes O,  Millett ER,  Nazareth I,  Nguyen-Van-Tam JS,  Watson JM,  Zambon M; Flu Watch Group, Johnson AM,  McMichael AJ. Natural T cell-mediated protection against seasonal and pandemic influenza. Results of the Flu Watch Cohort Study. Am J Respir Crit Care Med 2015; 191(12): 1422–1431

[65] van de Sandt CE,  Hillaire ML,  Geelhoed-Mieras MM,  Osterhaus AD,  Fouchier RA,  Rimmelzwaan GF. Human influenza A virus-specific CD8⁺ T-cell response is long-lived. J Infect Dis 2015; 212(1): 81–85

[66] Hillaire ML, Vogelzang-van Trierum  SE, Kreijtz JH,  de Mutsert G,  Fouchier RA,  Osterhaus AD,  Rimmelzwaan GF. Human T-cells directed to seasonal influenza A virus cross-react with 2009 pandemic influenza A (H1N1) and swine-origin triple-reassortant H3N2 influenza viruses. J Gen Virol 2013; 94(Pt 3): 583–592

[67] Valkenburg SA, Josephs  TM, Clemens EB,  Grant EJ,  Nguyen TH,  Wang GC,  Price DA,  Miller A,  Tong SY,  Thomas PG,  Doherty PC,  Rossjohn J,  Gras S, Kedzierska  K. Molecular basis for universal HLA-A*0201-restricted CD8⁺ T-cell immunity against influenza viruses. Proc Natl Acad Sci USA 2016; 113(16): 4440–4445

[68] Gras S, Kedzierski  L, Valkenburg SA,  Laurie K,  Liu YC, Denholm  JT, Richards MJ,  Rimmelzwaan GF,  Kelso A,  Doherty PC,  Turner SJ,  Rossjohn J,  Kedzierska K. Cross-reactive CD8⁺ T-cell immunity between the pandemic H1N1-2009 and H1N1-1918 influenza A viruses. Proc Natl Acad Sci USA 2010; 107(28): 12599–12604

[69] Quiñones-Parra S,  Grant E,  Loh L, Nguyen  TH, Campbell KA,  Tong SY,  Miller A,  Doherty PC,  Vijaykrishna D,  Rossjohn J,  Gras S, Kedzierska  K. Preexisting CD8⁺ T-cell immunity to the H7N9 influenza A virus varies across ethnicities. Proc Natl Acad Sci USA 2014; 111(3): 1049–1054

[70] van de Sandt CE,  Kreijtz JH,  de Mutsert G,  Geelhoed-Mieras MM,  Hillaire ML,  Vogelzang-van Trierum SE,  Osterhaus AD,  Fouchier RA,  Rimmelzwaan GF. Human cytotoxic T lymphocytes directed to seasonal influenza A viruses cross-react with the newly emerging H7N9 virus. J Virol 2014; 88(3): 1684–1693

[71] Kreijtz JH, de Mutsert  G, van Baalen CA,  Fouchier RA,  Osterhaus AD,  Rimmelzwaan GF. Cross-recognition of avian H5N1 influenza virus by human cytotoxic T-lymphocyte populations directed to human influenza A virus. J Virol 2008; 82(11): 5161–5166

[72] Lee LY, Ha  LA, Simmons C,  de Jong MD,  Chau NV,  Schumacher R,  Peng YC,  McMichael AJ,  Farrar JJ,  Smith GL,  Townsend AR,  Askonas BA,  Rowland-Jones S,  Dong T. Memory T cells established by seasonal human influenza A infection cross-react with avian influenza A (H5N1) in healthy individuals. J Clin Invest 2008; 118(10): 3478–3490

[73] van de Sandt CE,  Dou Y, Vogelzang-van Trierum  SE, Westgeest KB,  Pronk MR,  Osterhaus AD,  Fouchier RA,  Rimmelzwaan GF,  Hillaire ML. Influenza B virus-specific CD8⁺ T-lymphocytes strongly cross-react with viruses of the opposing influenza B lineage. J Gen Virol 2015; 96(8): 2061–2073

[74] Marsh SGE, Parham  P, Barber LD. The HLA factsbook. San Diego: Academic Press, 2000

[75] Berkhoff EGM, Boon  ACM, Nieuwkoop NJ,  Fouchier RAM,  Sintnicolaas K,  Osterhaus ADME,  Rimmelzwaan GF. A mutation in the HLA-B*2705-restricted NP383-391 epitope affects the human influenza A virus-specific cytotoxic T-lymphocyte response in vitro. J Virol 2004; 78(10): 5216–5222

[76] Boon ACM, de Mutsert  G, Graus YMF,  Fouchier RAM,  Sintnicolaas K,  Osterhaus ADME,  Rimmelzwaan GF. Sequence variation in a newly identified HLA-B35-restricted epitope in the influenza A virus nucleoprotein associated with escape from cytotoxic T lymphocytes. J Virol 2002; 76(5): 2567–2572

[77] Duan S, Thomas  PG. Balancing immune protection and immune pathology by CD8(+) T-cell responses to influenza infection. Front Immunol 2016; 7: 25

[78] Hillaire MLB, Rimmelzwaan  GF, Kreijtz JH. Clearance of influenza virus infections by T cells: risk of collateral damage? Curr Opin Virol 2013; 3(4): 430–437

[79] Román E, Miller  E, Harmsen A,  Wiley J,  Von Andrian UH,  Huston G,  Swain SL. CD4 effector T cell subsets in the response to influenza: heterogeneity, migration, and function. J Exp Med 2002; 196(7): 957–968

[80] Goodnow CC, Vinuesa  CG, Randall KL,  Mackay F,  Brink R. Control systems and decision making for antibody production. Nat Immunol 2010; 11(8): 681–688

[81] Ma CS, Deenick  EK, Batten M,  Tangye SG. The origins, function, and regulation of T follicular helper cells. J Exp Med 2012; 209(7): 1241–1253

[82] Belz GT, Wodarz  D, Diaz G,  Nowak MA,  Doherty PC. Compromised influenza virus-specific CD8(+)-T-cell memory in CD4(+)-T-cell-deficient mice. J Virol 2002; 76(23): 12388–12393

[83] Shedlock DJ, Shen  H. Requirement for CD4 T cell help in generating functional CD8 T cell memory. Science 2003; 300(5617): 337–339

[84] Morita R, Schmitt  N, Bentebibel SE,  Ranganathan R,  Bourdery L,  Zurawski G,  Foucat E,  Dullaers M,  Oh S, Sabzghabaei  N, Lavecchio EM,  Punaro M,  Pascual V,  Banchereau J,  Ueno H. Human blood CXCR5(+)CD4(+) T cells are counterparts of T follicular cells and contain specific subsets that differentially support antibody secretion. Immunity 2011; 34(1): 108–121

[85] Bentebibel SE, Lopez  S, Obermoser G,  Schmitt N,  Mueller C,  Harrod C,  Flano E,  Mejias A,  Albrecht RA,  Blankenship D,  Xu H, Pascual  V, Banchereau J,  Garcia-Sastre A,  Palucka AK,  Ramilo O,  Ueno H. Induction of ICOS+CXCR3+CXCR5+ TH cells correlates with antibody responses to influenza vaccination. Sci Transl Med 2013; 5(176): 176ra32

[86] Bentebibel SE, Khurana  S, Schmitt N,  Kurup P,  Mueller C,  Obermoser G,  Palucka AK,  Albrecht RA,  Garcia-Sastre A,  Golding H,  Ueno H. ICOS(+)PD-1(+)CXCR3(+) T follicular helper cells contribute to the generation of high-avidity antibodies following influenza vaccination. Sci Rep 2016; 6(1): 26494

[87] Herati RS, Muselman  A, Vella L,  Bengsch B,  Parkhouse K,  Del Alcazar D,  Kotzin J,  Doyle SA,  Tebas P,  Hensley SE,  Su LF, Schmader  KE, Wherry EJ. Successive annual influenza vaccination induces a recurrent oligoclonotypic memory response in circulating T follicular helper cells. Sci Immunol  2017; 2(8). pii: eaag2152 

[88] Bennett SRM, Carbone  FR, Karamalis F,  Flavell RA,  Miller JF,  Heath WR. Help for cytotoxic-T-cell responses is mediated by CD40 signalling. Nature 1998; 393(6684): 478–480

[89] Schoenberger SP, Toes  RE, van der Voort EI, Offringa R,  Melief CJ. T-cell help for cytotoxic T lymphocytes is mediated by CD40-CD40L interactions. Nature 1998; 393(6684): 480–483

[90] Johnson S, Zhan  Y, Sutherland RM,  Mount AM,  Bedoui S,  Brady JL,  Carrington EM,  Brown LE,  Belz GT,  Heath WR,  Lew AM. Selected Toll-like receptor ligands and viruses promote helper-independent cytotoxic T cell priming by upregulating CD40L on dendritic cells. Immunity 2009; 30(2): 218–227

[91] Tripp RA, Sarawar  SR, Doherty PC. Characteristics of the influenza virus-specific CD8⁺ T cell response in mice homozygous for disruption of the H-2lAb gene. J Immunol 1995; 155(6): 2955–2959

[92] Ahmed R, Butler  LD, Bhatti L. T4+ T helper cell function in vivo: differential requirement for induction of antiviral cytotoxic T-cell and antibody responses. J Virol 1988; 62(6): 2102–2106

[93] Seah SG, Carrington  EM, Ng WC,  Belz GT,  Brady JL,  Sutherland RM,  Hancock MS,  La Gruta NL,  Brown LE,  Turner SJ,  Zhan Y, Lew  AM. Unlike CD4+ T-cell help, CD28 costimulation is necessary for effective primary CD8⁺ T-cell influenza-specific immunity. Eur J Immunol 2012; 42(7): 1744–1754

[94] Dolfi DV, Duttagupta  PA, Boesteanu AC,  Mueller YM,  Oliai CH,  Borowski AB,  Katsikis PD. Dendritic cells and CD28 costimulation are required to sustain virus-specific CD8⁺ T cell responses during the effector phase in vivo. J Immunol 2011; 186(8): 4599–4608

[95] Olson MR, Seah  SG, Cullen J,  Greyer M,  Edenborough K,  Doherty PC,  Bedoui S,  Lew AM, Turner  SJ. Helping themselves: optimal virus-specific CD4 T cell responses require help via CD4 T cell licensing of dendritic cells. J Immunol 2014; 193(11): 5420–5433

[96] Sun JC, Bevan  MJ. Defective CD8 T cell memory following acute infection without CD4 T cell help. Science 2003; 300(5617): 339–342

[97] Laidlaw BJ, Zhang  N, Marshall HD,  Staron MM,  Guan T, Hu  Y, Cauley LS,  Craft J,  Kaech SM. CD4⁺ T cell help guides formation of CD103⁺ lung-resident memory CD8⁺ T cells during influenza viral infection. Immunity 2014; 41(4): 633–645

[98] McKinstry KK, Strutt  TM, Kuang Y,  Brown DM,  Sell S, Dutton  RW, Swain SL. Memory CD4⁺ T cells protect against influenza through multiple synergizing mechanisms. J Clin Invest 2012; 122(8): 2847–2856

[99] Marshall NB, Swain  SL. Cytotoxic CD4 T cells in antiviral immunity. J Biomed Biotechnol 2011; 2011: 954602

[100] Juno JA, van Bockel  D, Kent SJ,  Kelleher AD,  Zaunders JJ,  Munier CM. Cytotoxic CD4 T cells-friend or foe during viral infection? Front Immunol 2017; 8: 19

[101] Hua L, Yao  S, Pham D,  Jiang L,  Wright J,  Sawant D,  Dent AL,  Braciale TJ,  Kaplan MH,  Sun J. Cytokine-dependent induction of CD4⁺ T cells with cytotoxic potential during influenza virus infection. J Virol 2013; 87(21): 11884–11893

[102] Dutta A, Huang  CT, Lin CY,  Chen TC,  Lin YC, Chang  CS, He YC. Sterilizing immunity to influenza virus infection requires local antigen-specific T cell response in the lungs. Sci Rep 2016; 6(1): 32973

[103] Richards KA, Topham  D, Chaves FA,  Sant AJ. Cutting edge: CD4 T cells generated from encounter with seasonal influenza viruses and vaccines have broad protein specificity and can directly recognize naturally generated epitopes derived from the live pandemic H1N1 virus. J Immunol 2010; 185(9): 4998–5002

[104] Galli G, Medini  D, Borgogni E,  Zedda L,  Bardelli M,  Malzone C,  Nuti S, Tavarini  S, Sammicheli C,  Hilbert AK,  Brauer V,  Banzhoff A,  Rappuoli R,  Del Giudice G,  Castellino F. Adjuvanted H5N1 vaccine induces early CD4⁺ T cell response that predicts long-term persistence of protective antibody levels. Proc Natl Acad Sci USA 2009; 106(10): 3877–3882

[105] Zens KD, Chen  JK, Farber DL. Vaccine-generated lung tissue-resident memory T cells provide heterosubtypic protection to influenza infection. JCI Insight 2016; 1(10): e85832

[106] He XS, Holmes  TH, Zhang C,  Mahmood K,  Kemble GW,  Lewis DB,  Dekker CL,  Greenberg HB,  Arvin AM. Cellular immune responses in children and adults receiving inactivated or live attenuated  influenza  vaccines.  J Virol 2006; 80(23): 11756–11766

[107] Peng Y, Wang  B, Talaat K,  Karron R,  Powell TJ,  Zeng H, Dong  D, Luke CJ,  McMichael A,  Subbarao K,  Dong T. Boosted influenza-specific T cell responses after H5N1 pandemic live attenuated influenza virus vaccination. Front Immunol 2015; 6: 287

[108] Wang Z, Chua  BY, Ramos JV,  Parra SM,  Fairmaid E,  Brown LE,  Jackson DC,  Kedzierska K. Establishment of functional influenza virus-specific CD8(+) T cell memory pools after intramuscular immunization. Vaccine 2015; 33(39): 5148–5154

[109] Berthoud TK, Hamill  M, Lillie PJ,  Hwenda L,  Collins KA,  Ewer KJ,  Milicic A,  Poyntz HC,  Lambe T,  Fletcher HA,  Hill AV,  Gilbert SC. Potent CD8⁺ T-cell immunogenicity in humans of a novel heterosubtypic influenza A vaccine, MVA-NP+M1. Clin Infect Dis 2011; 52(1): 1–7

[110] Powell TJ, Peng  Y, Berthoud TK,  Blais ME,  Lillie PJ,  Hill AV,  Rowland-Jones SL,  McMichael AJ,  Gilbert SC,  Dong T. Examination of influenza specific T cell responses after influenza virus challenge in individuals vaccinated with MVA-NP+M1 vaccine. PLoS One 2013; 8(5): e62778

[111] Lillie PJ, Berthoud  TK, Powell TJ,  Lambe T,  Mullarkey C,  Spencer AJ,  Hamill M,  Peng Y, Blais  ME, Duncan CJ,  Sheehy SH,  Havelock T,  Faust SN,  Williams RL,  Gilbert A,  Oxford J,  Dong T, Hill  AV, Gilbert SC. Preliminary assessment of the efficacy of a T-cell-based influenza vaccine, MVA-NP+M1, in humans. Clin Infect Dis 2012; 55(1): 19–25

[112] Pleguezuelos O, Robinson  S, Stoloff GA,  Caparrós-Wanderley W. Synthetic Influenza vaccine (FLU-v) stimulates cell mediated immunity in a double-blind, randomised, placebo-controlled Phase I trial. Vaccine 2012; 30(31): 4655–4660

[113] Francis JN, Bunce  CJ, Horlock C,  Watson JM,  Warrington SJ,  Georges B,  Brown CB. A novel peptide-based pan-influenza A vaccine: a double blind, randomised clinical trial of immunogenicity and safety. Vaccine 2015; 33(2): 396–402

[114] Atsmon J, Kate-Ilovitz  E, Shaikevich D,  Singer Y,  Volokhov I,  Haim KY,  Ben-Yedidia T. Safety and immunogenicity of multimeric-001— a novel universal influenza vaccine. J Clin Immunol 2012; 32(3): 595–603

[115] Pleguezuelos O, Robinson  S, Fernández A,  Stoloff GA,  Mann A, Gilbert  A, Balaratnam G,  Wilkinson T,  Lambkin-Williams R,  Oxford J,  Caparrós-Wanderley W. A synthetic influenza virus vaccine induces a cellular immune response that correlates with reduction in symptomatology and virus shedding in a randomized phase Ib live-virus challenge in humans. Clin Vaccine Immunol 2015; 22(7): 828–835

[116] Kopecky-Bromberg SA,  Fraser KA,  Pica N, Carnero  E, Moran TM,  Franck RW,  Tsuji M,  Palese P. Alpha-C-galactosylceramide as an adjuvant for a live attenuated influenza virus vaccine. Vaccine 2009; 27(28): 3766–3774

RIGHTS & PERMISSIONS

Higher Education Press and Springer-Verlag GmbH Germany, part of Springer Nature