How to assess the effectiveness of nasal influenza vaccines? Role and measurement of sIgA in mucosal secretions

Abstract Secretory IgAs (sIgA) constitute the principal isotype of antibodies present in nasal and mucosal secretions. They are secreted by plasma cells adjacent to the mucosal epithelial cells, the site where infection occurs, and are the main humoral mediator of mucosal immunity. Mucosally delivered vaccines, such as live attenuated influenza vaccine (LAIV), are able to mimic natural infection without causing disease or virus transmission and mainly elicit a local immune response. The measurement of sIgA concentrations in nasal swab/wash and saliva samples is therefore a valuable tool for evaluating their role in the effectiveness of such vaccines. Here, we describe two standardized assays (enzyme‐linked immunosorbent assay and microneutralization) available for the quantification of sIgA and discuss the advantages and limitations of their use.

"backup" antibodies whose function is to respond in case of systemic infection due to invasion across the mucosal epithelium. 4 IgG-secreting cells are produced in the mucosa-associated lymphoid tissues (MALTs) and regional lymph nodes. IgG antibodies are secreted in the bloodstream and, reached the mucosal tissues, move via diffusion from the serum to the mucus. 5

| Role of cellular immune responses
Along with the humoral immunity, also the cell-mediated immune (CMI) response is activated after influenza infection. Unlike humoral response, capable of neutralizing activity, CMI is able to prevent virus replication and decrease the time for recovery. 7,8 CD4 + follicular helper T (Th) lymphocytes in presence of antigen-presenting cells (APCs), such as dendritic cells (DCs) and influenza antigens, induce the differentiation of naïve B cells into IgA-secreting plasma cells (PCs). sIgA constitute the principal isotype of antibodies present in external secretions, such as nasal fluid, saliva, milk, colostrum intestinal fluid, and gallbladder bile. 9 In the upper respiratory tract, sIgA antibodies are secreted by mucosal PCs adjacent to the mucosal epithelial layer at the site of infection 10 and represent the main humoral mediator of nasal immunity. 11

| Immune mechanisms contributing to disease reduction or protection
In influenza-naive subjects, the clearance of primary viral infection occurs through sIgA and cytotoxic T lymphocytes (CTLs). More specifically, sIgA appear on day 5 post-infection, and their level rapidly increases in the nasal wash until day 7-10 post-infection, when it reaches a plateau. IgA local immune response persists for a period of 3-5 months 5,12,13 and then gradually diminishes returning to the pre-immunization levels within 6 months. 14 In addition, it is possible to detect IgA-producing memory cells locally. 5,12,13 CTLs appear transiently in the nasal mucosa and peak on day 7 after infection.
sIgA have a pivotal role in protecting against influenza infection of the upper respiratory mucosal surfaces, since they can disarm the virus either before it crosses the mucosal barrier 15 or in infected epithelial cells by intracellular neutralization. 15,16 The magnitude of the IgA antibody response is directly correlated with resistance to new infections. 17 In addition, IgA is the predominant Ig isotype in local secretions after secondary infection and an IgA response is also detected in the serum upon subsequent infection which support its additional important role in protection against influenza virus re-infections. 18 Along with IgA, IgM antibodies are also secreted actively across the mucosa and may contribute to protection by preventing viral entry into the cells and also interfering with virus replication in the cells. 5,18 The potential protective role of IgM antibodies is supported by a study in mice which has shown that IgM antibodies can

| IgA immune responses upon influenza virus infection of the mucosa of the upper respiratory tract
On the basolateral surface of the epithelial cells in the lamina propria of mucosal tissue, a polymeric Ig receptor (pIgR) links the dimeric IgA (dIgA) and moves to the apical side ( Figure 1). During the pro- The presence of SC provides sIgA a greater functional stability, both by masking the protease sites from proteolytic degradation operated by proteases present in mucosal secretions 21 and by sustaining the association of monomeric IgA (mIgA). 22 IgA do not promote the activation of the inflammatory complement system, a feature which is critical to maintaining the integrity of the mucosal barrier. 23

| Presentation forms and functions of IgA antibodies
In human serum, IgA are mainly present in the monomeric form with two α-heavy and two light chains. On the other hand, in external secretions IgA are highly heterogeneous in terms of their quaternary structure but the majority are in polymeric form. 24 sIgA are generally present as a dimer, despite, and at low frequency, as larger polymeric forms (pIgA) especially tetramers. 25 It has been hypothesized that pIgA may have a higher ability than mIgA to neutralize intracellular viral particle assembly by binding newly synthesized viral proteins. 5,17,26 It has also been demonstrated that the polymeric nature of sIgA was responsible for their elevated cross-reactivity, thereby increasing the avidity of this antibody subclass in comparison with mIgA and serum IgG. 27,28 The best neutralizing activity and the higher avidity of human pIgA than mIgA can be attributed to the presence of multiple antigen-binding sites located on each Ig polymer, indicating that the quaternary structure plays a key role for their potency. 25 This result is in accordance with previous researches conducted on mice. 27,29,30 A recent study by Saito and colleagues demonstrated that IgA tetramerization improves target breadth exerting no effect on potency of functionality of anti-influenza virus broadly neutralizing antibody. 31 The higher anti-viral activity of pIgA than mIgA is particularly important, considering the anatomical site of sIgA action. 10 pIgA appears to have a greater inhibitory potential in preventing viral attachment and virus neutralization than mIgA and also IgG. 29,32,33 Another study showed the existence of larger pIgA in addition to tetrameric sIgA in the upper respiratory tract.
The proportion of this polymeric form is approximately 20% of the total IgA. 34 In summary, the mucosal surface is endowed with two protective barriers against viral infection, both of which involve mucosal IgA, that is, extracellular sIgA and intracellular pIgA. 29

| I G A IMMUNE RE S P ON S E UP ON INFLUENZ A VACCINATION
Conventional inactivated influenza vaccines (IIVs), generally delivered through subcutaneous or intramuscular injection, are today still the most efficient, valuable, and low-cost tools to effectively reduce influenza virus infections and subsequently morbidity and mortality. 35 This parental administration is able to increase the serum antibody level in the systemic immune compartment, but it is not able to trigger a local mucosal immune response at the site of primary infection, that is, an induction of sIgA which exhibit a wide cross-protection activity. This represents a limit for conventional inactivated influenza vaccines in conferring full protection against infection. 36 While natural infection is able to induce both mucosal and systemic heterosubtypic responses, the immunity induced by parenterally application of inactivated influenza vaccines is generally virus subtype-specific. 37 In pre-immunized subjects, the natural contact Studies performed in mice have demonstrated the predominant protective role played by sIgA, 43,44 even in case of absence of T cells. 45 Specifically, the passive intranasal transfer of anti-influenza A IgA from the respiratory tract of mice immunized with live influenza virus has been seen to provide protection in naive mice. 43 Accordingly, this protection was suppressed by the intranasal instillation of anti-IgA, 46 whereas it was not affected by treatment with anti-IgM or anti-IgG antibodies. This result supports the importance of IgA as a mediator of murine nasal anti-influenza virus immunity in immunocompetent mice. 47 Furthermore, several studies have found a higher level of correlation between the degree of protection and the antibody secretory level than serum antibodies both in mice 48

| ELISA A SSAYS FOR THE DE TERMINATI ON OF THE I G A CONTENT IN TE S T SAMPLE S
According to the type of influenza vaccine used and the route of

| Influenza HA-Specific IgA respect to Total IgA
This method is used to normalize the influenza-specific IgA content of a sample through the total IgA content ( Figure 2). The total IgA concentration in nasal wash/swab samples or saliva can easily be measured by using one of the many standardized ELISA kits available on the market. Concerning influenza-specific IgA detection, the procedure needs to be adapted due to the absence of a standardized human influenza-specific IgA reference. The ELISA procedure in principle has been described elsewhere. 73  important to run multiple samples collected from the same subject at different time points in the same ELISA plate.14 According to this method, the value of influenza-specific IgA normalized through the total IgA content will be expressed as "(Influenza-Specific IgA (U/mL)/ Total IgA (µg/mL)) * 100.".

| Influenza HA-Specific IgA and Total Protein
The basis of this method is the measurement of the total protein content in the samples (Figure 2). The determination of influenza HA-specific IgA with respect to total protein generally proves to be the best choice when a large number of samples have to be evaluated, since it is easy to use, sensitive, and rapid. 75 Depending on the total protein concentration obtained, two different methods of calculations can be adopted (Figure 2).
The first method will be applied if the total protein of the samples is higher than or equal to 1 mg/mL by standardization of nasal or saliva samples to a defined total protein content ( Figure 2) which may vary according to the type of sample (nasal wash vs nasal swab vs saliva). The influenza-specific IgA antibody titer is then calculated as the reciprocal of the highest dilution that yields an OD signal greater than or equal to a predefined cutoff value. However, since completely negative human nasal samples are usually not available, the exact calculation of a cutoff may not be optimal and require alternative approaches in the future.
One approach may be to use the "limit of blank" according to the following formula: "Average of background signals (OD Blank ) plus 2 standard deviations." 72 In this case, the cutoff value will be calculated without the need for a specific human sample; only ELISA reagents will be added to the coated plate together with the influenza antigen, and the background signal will be used to calculate the cutoff. An alternative possibility is the calculation of the cutoff value as the reciprocal of the highest dilution that shows an absorbance value >0.2 of the OD value after subtraction of the background as previously described. 73 In the case of low total protein concentrations of the samples in general or of big differences of the total protein content of different test samples, an alternative approach, combining the two approaches described above, can be used which is based on the estimation of IgA content by using the ratio between the titer and the total IgA content (Figure 2).

| NEUTR ALIZING (NT ) ANTIBOD IE S IN NA SAL WA S H/SWAB AND FUTURE A SSAYS
Some recent studies 74,76 have assessed neutralization (NT) antibody levels in standardized nasal wash/swab samples after intranasal immunization since NT antibodies are generally considered more specific than hemagglutination inhibition (HI) antibody titers in children vaccinated with LAIVs. However, it has been shown in a previous pediatric study with LAIV that influenza virus-specific salivary IgA levels correlated with serum HI responses, 76 although it is also discussed that the measured HI titers may underestimate the protective potential of LAIVs. 60,77 NT antibodies in serum samples are usually assessed by means of the microneutralization (MN), either CPE (cytopathic effect)-based 74 or ELISA-based, 37 or the plaque-reduction neutralization (PRNT) assay. In the present review, we focused on the CPE-based MN assay, since this is the preferred method because of its simplicity of execution, its ability to evaluate large numbers of samples, and the standardization of the quantity of virus used in the assay. 78 Along with the ELISA sIgA assay, the MN assay constitutes a valid approach to evaluate the immunogenicity of LAIVs, IIVs, or recombinant influenza vaccines (eg rHA) in inducing selective anti-influenza antibodies with influenza virus-neutralizing potential.
Beside classical ELISA-based and NT assays specific anti-HA influenza antibodies, there are newer assays with increased precision and sensitivity, such as the XMAP (x = analyte MAP = Multi-analite profiling) technology adapted for Luminex-based IgA assays. 79,80 The XMAP technology is a serological method that can be applied to

| CON CLUS IONS
Influenza vaccines elicit protective immunity before a new influenza virus variant is able to spread; they therefore constitute a primary protection tool. Although the main protective effectors against influenza virus infection are CTLs, IgG, and IgA located in the respiratory mucosa, most of the vaccines currently available are inactivated vaccines that are administered via parenteral injection, and which mostly promote serum IgG rather than mucosal IgA (rev. in [82][83][84] ). The importance of intranasally applied LAIVs is their ability to reproduce a natural infection without causing disease or virus transmission.
They mimic the natural encounter with the antigen by activating the innate immune system and promoting antibody and T cell-mediated immune responses. This type of vaccine can induce a broader immune response in children than intramuscular vaccines. 31,85,86 Furthermore, mucosal vaccines can elicit cross-reactive antibodies in humans. However, the development of cross-protective T lymphocytes has been observed in animal models, but this has not yet confirmed in humans. 85,86 An additional advantage of LAIVs is their consumer-friendly needle-free intranasal application which represents a minimal invasive delivery method, and it is expected with higher production capacities and a more widely distribution. For these reasons, its expanded use could increase the influenza vaccination coverage globally.
Furthermore, it may represent a favorable approach for mass immunizations, especially in younger children since its application is not associated with pain. 87 Although LAIVs have been on the global market for many years, no established correlates of protection for them are yet available. 77 Moreover, previously reported discrepancies of efficacy data from Europe and the US further complicate the understanding of the immune response elicited by LAIV. 77 Despite these complications, great efforts have been made in the recent years to develop novel intranasally administered vaccines to promote influenza virus-specific sIgA, 30,88 which, as has been widely reported, provide broader protection than serum IgG. A robust mucosal response is fundamental in order to protect both the single individual and the entire population by preventing transmission of the virus to susceptible subjects. 89 Notably, the use of the ELISA assay for IgA detection could play a major role in the evaluation of vaccine efficacy or effectiveness in the field, as currently influenza vaccine efficacy is traditionally assessed by means of serological assays that detect influenza-specific serum antibodies induced by the vaccine itself. However, these assays cannot be properly applied to intranasal vaccines, which mainly induce local immune responses (rev. in 90 ).
In conclusion, the measurement of sIgA in mucosal secretions for the evaluation of influenza vaccine efficacy or effectiveness and, in addition, also of the effectiveness of vaccines against other respiratory virus infections of the respiratory mucosae, is arousing great interest and may constitute a valuable asset.

CO N FLI C T O F I NTE R E S T
There are no conflicts of interest in the conduction of this study.

AUTH O R CO NTR I B UTI O N S
EG and A.M involved in writing, reviewing, and editing processes and prepared the images; O. K., CT, and I.M involved in reviewing and editing processes; and EM involved in supervision and review process.