FIELD OF THE INVENTION

The present invention relates to vaccine compositions comprising at least three variants of a subtype of an influenza virus antigen, wherein the at least three variants are selected on the basis of differences between their amino acid sequences. The present invention further relates to said vaccine compositions for use in a method of generating an immune response against influenza viruses. Further provided is a method for producing said vaccine compositions. BACKGROUND

Human influenza (human flu) is a highly contagious respiratory disease typically starting with an abrupt onset of fever, sore throat, blocked or running nose, headache, photophobia, dry cough and malaise. It gives rise to repeating and frequent epidemics and pandemics that occur suddenly, causing substantial morbidity and mortality. The first recorded influenza pandemic dates back to 1580. Over the course of history there have been several influenza pandemics that have sickened and killed millions. Most cases of death have been found to be a result of an increased physiologic load in an already compromised host, or to be the outcome of the combined effects of the viral disease and a secondary bacterial infection. The 1918 influenza virus, called the "Spanish flu", was particularly lethal, accounting for more than 40 million deaths worldwide and this was when rapid air travel was much less common. Albeit this strain caused pneumonia, also in this pandemic most deaths were associated with secondary bacterial pathogens.

Influenza viruses are RNA viruses that replicate their genome in the nucleus of the host cell. They belong to the family Orthomyxoviridae and are divided into three genera A, B and C, which can be distinguished by antigenic differences in two of the structural proteins of the virus, the matrix protein M2 and the nucleoprotein. Each of these types has many strains. These are enveloped viruses with a segmented genome containing seven or eight single- stranded segments of negative-sense RNA. Each of these RNA segments contains one or two genes. The genomes of influenza A and influenza B virus consist of eight RNA segments, which are coding for 12 viral proteins (Steinhauer, D.A. & Skehel, J.J., Ann. Rev. Genet. (2002), 36, 305-332; Hutchinson, E.C., et al., Journal of General Virology (2010) 91 , 313-328). The three largest gene segments of influenza A virus encode the subunits of the viral polymerase, PB2, PB1 , and PA. The fourth segment encodes the hemagglutinin glycoprotein (HA), responsible for binding to cell-surface receptors and membrane fusion, and the fifth gene segment encodes the nucleoprotein (NP), which encapsidates cRNAs and vRNAs, which allows them to be recognized as templates for the viral polymerase. Segment 6 encodes the neuraminidase (NA), which cleaves sialic acid from virus and host cell glycoconjugates to allow mature virus particles to be released. The seventh segment generates two gene products, the matrix protein, M1 , and the M2 transmembrane protein, which has proton channel activity. In influenza B virus this segment encodes matrix protein M1 and BM2, thought to be a functional counterpart of M2. The eighth gene segment encodes the protein NS1 , which inter alia sequesters ds RNA formed during virus replication, and the nuclear export protein (NEP). In order to produce an intact virion or infectious influenza A virus an effective incorporation of all 8 gene segments into a viral particle is necessary.

Given the fact that the hemagglutinin surface proteins (HA) exist in 18 subtypes, the neuraminidase (NA) in 1 1 subtypes, and the potential for recombination exists in the animal kingdom as well as in the human, many potential pandemic influenza A virus candidates exist. Once an influenza virus has become a seasonal virus, usually after a pandemic, it is going to drift or change over time. Several seasonal viruses are presently in co-circulation: An influenza A virus of the subtype H3N2, and another of the subtype H1 N1 , and two influenza virus type B strains from the Yamagata and the Victoria lineages. After the recent swine flu pandemic, the new variant of the influenza A H1 N1 subtype (vH1 N1 or H1 N1 new) became the new seasonal H1 N1 strain.

Influenza B and C viruses can infect only humans, although there have been reports of influenza B virus isolation from seals and influenza C virus isolation from pigs. In contrast thereto, Influenza A viruses can infect both mammals and birds. The most devastating flu viruses of the 20th century, the Spanish flu pandemic in 1918 (H1 N1 ), the Asian flu pandemic in 1957 (H2N2) and the Hong Kong flu pandemic in 1968 (H3N2), were all of avian origin. Aquatic birds are natural reservoirs of influenza A viruses. These viruses are known to cross the species barrier and cause either transitory infections or establish permanent lineages in mammals including man. While influenza B viruses do not have pandemic potential, they cause significant disease and are the predominant circulating strain of influenza virus approximately one in every 3 years. Influenza B virus is therefore an essential component of the influenza vaccine administered to susceptible groups such as the elderly and asthmatic. In any case, approaches for pandemic influenza vaccines, as well as seasonal influenza vaccines, warrant a combination of several influenza viruses. For pre-pandemic vaccines, a combination of several strains into one vaccine candidate is indicated to either prime against several viruses simultaneously in a pre-pandemic setting or to limit a stockpile to a few vaccines (vaccine library), but each vaccine with a multivalent option, i.e. being protective against several strains to increase its potential. At present, seasonal vaccines should cover at least three strains, two A-strains and one B-strain.

In the case of a pandemic vaccine, generally the entire population should be administered with the vaccine, while in the case of a seasonal vaccine, primarily risk groups such as young children and the elderly should be administered with the vaccine.

For priming in naive populations an induction of immunity as similar as possible to the wild type virus infection in regard to internal and external antigens is desired. For boosting vaccination in subjects already primed by either a wild type influenza (sometimes also called "flu" herein) infection or a flu vaccination, pre-existing immunity should not prohibit a sufficient booster response. Priming can consist of one or several doses (a priming schedule) and boosting most often of only one vaccination. The principal mechanism of action of current subunit or inactivated, detergent-disrupted influenza virus vaccines is to induce neutralizing antibodies (Doherty et al. 2008, The Journal of Clinical Investigation 1 18, 3273-3275). Commonly used inactivated seasonal influenza vaccines induce protective antibody responses against the immunizing virus strains (Brown et al. 2009, Immunology and Cell Biology 87, 300-308). However, the antibody response may not be effective against novel virus strains. Antigenic drift occurs in both type A and type B influenza and results in neutralization-resistant mutants. Id. Thus, it is necessary to constantly produce new vaccines to combat these new strains.

On the one hand, the current practice of annual reformulation of influenza vaccines is highly cost-intensive, both at the level of production and distribution/administration of the vaccine. On the other hand, reformulated vaccines might become available only after a novel influenza strain has started spreading in a population. Accordingly, there is a great need to improve current influenza vaccines. For novel vaccine compositions it would be desired that they provide vaccinated subjects with broad protection against influenza strains both circulating at the time of vaccination and predicted to arise in the future, e.g., from antigenic drift or antigenic shift. Subtypes of influenza virus antigens, such as HA or NA, have variants and these variants are rather polymorph, i.e., they differ among each other in their amino acid sequence. All the more, these already polymorphic variants underlie the phenomenon of antigenic drift and antigenic shift. While influenza viruses are changing by antigenic drift all the time, antigenic shift happens only occasionally.

In antigenic drift small changes in, e.g. the HA gene of influenza viruses happen continually over time as the virus replicates. These small genetic changes usually produce viruses that are rather closely related to one another. However, these small genetic changes can accumulate over time and result in viruses that are antigenically different (further away on the phylogenetic tree). When this happens, the immune system may no longer recognize those viruses. This is also why influenza vaccine composition must be reviewed each year, and updated as needed to keep up with evolving viruses. Antigenic shift is an abrupt, major change in the influenza viruses, resulting in new HA and/or NA proteins in influenza viruses. Shift results in a new influenza subtype or a virus with a hemagglutinin or a hemagglutinin and neuraminidase combination that has emerged from an animal population that is so different from the same subtype in humans that most people do not have immunity to the new (e.g. novel) virus. Such a "shift" occurred in the spring of 2009, when an H1 N1 virus with a new combination of genes emerged to infect people and quickly spread, causing a pandemic. When shift happens, most people have little or no protection against the new virus.

Huber et al. (2009), Vaccine 27, 1 192-122 have shown that a prime with DNA followed by a live attenuated virus boost with mixture of three H3N2 viruses (HK68, VI75, LE86) covers 20 years of antigenic drift in mice. Carter et al. (2009), J Virol 87, 1400-1410 show that sequential infection in ferrets with three seasonal H1 N1 strains (either historical or modern) confers protection against H1 N1 pdm challenge (not contained in infection series). Moreover a mixture of sera obtained after infection with single viruses neutralised H1 N1 pdm, while none of the separate sera had neutralising activity. Prabakaran et al. (2014), PLos One Sep 17;9(9):e107316 show that three MVA-expressed H5 antigens induce broad cross-clade neutralising responses in mice and guinea pigs. A recent publication by Schwartzman et al. (2015), MBio. Jul-Aug; 6(4), e01044-15 suggest that HA group 1 / 2 transcending immunity can be achieved in mice by intranasal immunisation with a mixture of H1 , H3, H5 and H7 VLPs. Mice were protected from lethal challenge with a number of influenza A viruses not contained in the vaccine. From these approaches, it is, however, not clear which and how many separate components would be required to adequately cover all influenza A haemagglutinins. In view of the issue with antigenic drift of polymorphic variants of a subtype of an influenza virus resulting in variants which are no longer recognized by the immune system, the technical problem underlying the present invention is to provide influenza vaccine compositions which are able to induce (cross)protection against influenza virus infections due to increased antibody breadth, yielding protection not only to the influenza variants of a subtype of an influenza virus antigen represented in the vaccine compositions, but also to variants of said subtype not included in the vaccine composition.

SUMMARY OF THE INVENTION

The present invention provides a solution to the technical problem of providing novel influenza vaccine compositions which are able to induce protection against influenza virus infections, wherein protection is not restricted to the specific variants of an influenza subtype represented in the vaccine compositions, but also to variants not included in the vaccine composition.

Basically, the present inventor found that immunization with a mixture of variants of a subtype of an influenza virus antigen yielded functional antibody levels to all variants comparable to levels induced by monovalent immunization and also to variants not included in the vaccine composition. The mechanism behind the observed broadening was shown to be an increase in the fraction of cross-reactive antibodies, most likely because variant- specific epitopes are present at lower frequency relative to conserved epitopes. In other words, it is assumed that the broadening of the antibody response is due to the increased relative concentration of common epitopes diluting out variant specific epitopes. The present inventor thereby developed the so-called epitope dilution phenomenon (EDiP) as a practical strategy for the induction of broad, cross-variant antibody responses against polymorphic antigens.

In practice, the present inventor developed certain rules for the choice of variants of a subtype of an influenza virus antigen, such as HA or NA, with the aim of "educating" the immune system by a combinatorial immunization strategy to be able to recognize a broad range of variants of a subtype of an influenza virus antigen which also includes potential new subtypes.

In fact by following his rules on the choice of variants, the present invention demonstrates for influenza virus H1 , H3 and B strains that the immune response was broadened beyond the variants of the subtype included in a vaccine composition, i.e. strains isolated -10 years after the last strain included in the vaccine are neutralised by mouse sera. For H1 strains, the following HA variants were used:

A/Puerto Rico/8/1934 (SEQ ID No: 1 ),

A/USSR/92/1977 (SEQ ID No: 2),

A/Texas/36/1991 (SEQ ID No: 3),

A/New Caledonia/20/1999 (SEQ ID No: 4), and

A/Brisbane/59/2007 (SEQ ID No: 5).

A broadening against A California/04/2009 (SEQ ID No: 6) or A/California/07/2009 (SEQ ID No: 7) or A/Mexico/INDRE4487/2009 (SEQ ID No: 8) can be expected. It may, however, not yet be measurable, since is assumed that the micro neutralization (MN) assay may lack sensitivity to detect A/California/4/2009 and/or A/Mexico/INDRE4487/2009 responses in mouse serum as has been previously observed with mini HA stem antigen constructs (Impagliazzo et al. (2015), Science 349(6254), 101-106).

For H3 strains, the following HA variants were used:

A/Hong Kong/1/1968 (SEQ ID No: 9),

A/England/321/1977 (SEQ ID No: 10),

A/Sichuan/2/1987 (SEQ ID No: 1 1 ),

A/Nanchang/933/1995 (SEQ ID No: 12) or A/Johannesburg/33/94 (SEQ ID No: 13), and A/Wyoming/3/2003 (SEQ ID NO: 14) or A/Fujian/41 1/2002 (SEQ ID No: 15).

A broadening to A/Switzerland/9715293/2013 (SEQ ID No: 16) and/or A Hong Kong/4801/2014 (SEQ ID No: 17, 18, 19 or 20) was surprisingly observed.

For B strains, the following HA variants were used:

B/Hong Kong/8/73 (SEQ ID No: 21 ),

B/Beijing/1/1987 (SEQ ID No: 22) or B/Victori a/02/1987 (SEQ ID No: 23),

B/Yamagata/16/1988 (SEQ ID No: 24),

B/Jiangsu/10/2003 (SEQ ID No: 25), and

B/Malaysia/2506/2004 (SEQ ID No: 26).

A broadening to B/Brisbane/60/2008 (SEQ ID No: 27 or 28) and/or B/Phuket/3073/2013 (SEQ ID No: 29 or 30) was surprisingly observed.

In sum, the inventor of the present application found that it is possible to broaden the scope of protection conferred to a subject upon vaccination compared to the scope of protection conferred by current influenza vaccines. To this end, a specific combination of variants of a subtype of an influenza virus antigen is provided in an influenza vaccine composition, said specific combination is based on the rules for choosing variants as provided herein. Immunity conferred by the vaccine composition of the present invention is advantageously not limited to variants which are actually represented in the vaccine composition. Rather, the vaccine composition of the present invention also confers immunity against other variants which are not included in the vaccine composition. These variants could have been circulating before or at the time point of vaccination of a subject. Importantly, the vaccine composition of the present invention may even confer protection against future variants which have not yet been circulating at the time of providing or administering the vaccine composition. Such future influenza virus strains might arise, e.g., from antigenic shift or antigenic drift.

Accordingly, the present invention provides a vaccine composition comprising

(a) at least three variants of a subtype of an influenza A virus antigen, wherein

(i) a first and second variant of a subtype of an influenza A virus antigen differ from each other by at least 10 amino acids over their entire length;

(ii) a second and third variant of a subtype of an influenza A virus antigen differ from each other by at least 10 amino acids over their entire length; and (iii) a first and third variant of a subtype of an influenza A virus antigen differ from each other by more than 35 amino acids over their entire length; and/or (b) at least three variants of a subtype of an influenza B virus antigen, wherein

(i) a first and second variant of a subtype of an influenza B virus antigen differ from each other by at least 10 amino acids over their entire length;

(ii) a second and third variant of a subtype of an influenza B virus antigen differ from each other by at least 10 amino acids over their entire length; and

(iii) a first and third variant of a subtype of an influenza B virus antigen differ from each other by more than 20 amino acids over their entire length. However, preferably the first and second variant of a subtype of an influenza A virus antigen differ from each other by no more than 60, 55, 50, 45, 40, 35, or 30 amino acids over their entire length.

However, preferably the first and third variant of a subtype of an influenza A virus antigen differ from each other by no more than 70, 65, 60, 55, or 50 amino acids over their entire length.

However, preferably the first and second variant of a subtype of an influenza B virus antigen differ from each other by no more than 60, 55, 50, 45, 40, 35, or 30 amino acids over their entire length.

However, preferably the first and third variant of a subtype of an influenza B virus antigen differ from each other by no more than 70, 65, 60, 55, 50, 45, 40 or 35 amino acids over their entire length. The present invention also provides a vaccine composition as described herein for use in a method of generating an immune response against influenza virus A and/or B.

Furthermore, the present invention provides a vaccine composition for use in a method of effecting cross-neutralization against at least one variant of said subtype of said influenza virus A and/or B antigen which is different from said variants comprised in said vaccine.

Also, the present invention provides the use of at least three variants of a subtype of an influenza A virus antigen, wherein

(i) a first and second variant of a subtype of an influenza A virus antigen differ from each other by at least 10 amino acids over their entire length;

(ii) a second and third variant of a subtype of an influenza A virus antigen differ from each other by at least 10 amino acids over their entire length; and

(iii) a first and third variant of a subtype of an influenza A virus antigen differ from each other by more than 35 amino acids over their entire length

for the manufacture of a vaccine composition which effects cross-neutralization against at least one variant of said subtype of said influenza A virus antigen which is different from said variants comprised in said vaccine. Moreover, the present invention provides the use of at least three variants of a subtype of an influenza B virus antigen, wherein

(i) a first and second variant of a subtype of an influenza B virus antigen differ from each other by at least 10 amino acids over their entire length;

(ii) a second and third variant of a subtype of an influenza B virus antigen differ from each other by at least 10 amino acids over their entire length; and

(iii) a first and third variant of a subtype of an influenza B virus antigen differ from each other by more than 20 amino acids over their entire length for the manufacture of a vaccine composition which effects cross-neutralization against at least one variant of said subtype of said influenza B virus antigen which is different from said variants comprised in said vaccine.

The present invention provides a method for producing a vaccine composition against influenza A virus, comprising

(a) selecting at least three variants of a subtype of a variant of an influenza A virus antigen, wherein

(i) a first and second variant of a subtype of an influenza A virus antigen differ from each other by at least 10 amino acids over their entire length; (ii) a second and third variant of a subtype of an influenza A virus antigen differ from each other by at least 10 amino acids over their entire length; and

(iii) a first and third variant of a subtype of an influenza A virus antigen differ from each other by more than 35 amino acids over their entire length

(b) combining said selected variants in one vaccine composition; and

(c) optionally determining whether said vaccine composition effects cross-neutralization against at least one variant of said subtype of said influenza A virus antigen which is different from said variants comprised in said vaccine.

The present invention provides a method for producing a vaccine composition against influenza B virus, comprising

(a) selecting at least three variants of a subtype of a variant of an influenza B virus antigen, wherein

(i) a first and second variant of a subtype of an influenza B virus antigen differ from each other by at least 10 amino acids over their entire length;

(ii) a second and third variant of a subtype of an influenza B virus antigen differ from each other by at least 10 amino acids over their entire length; and

(iii) a first and third variant of a subtype of an influenza B virus antigen differ from each other by more than 20 amino acids over their entire length

(b) combining said selected variants in one vaccine composition; and

(c) optionally determining whether said vaccine composition effects cross-neutralization against at least one variant of said subtype of said influenza B virus antigen which is different from said variants comprised in said vaccine. BRIEF DESCRIPTION OF THE FIGURES

Figure 1 : H1 N1 neutralisation titres; see also Table 1 below for an explanation of the abbreviations used in the Figure

Figure 2: H1 N1 IgG titres; see also Table 2 below for an explanation of the

abbreviations used in the Figure

Figure 3: H3N2 neutralisation titres; see also Table 3 below for an explanation of the abbreviations used in the Figure

Figure 4: H3N2 IgG titres; see also Table 4 below for an explanation of the

abbreviations used in the Figure

Figure 5: B strain neutralisation titres; see also Table 5 below for an explanation of the abbreviations used in the Figure

Figure 6: B strain IgG titres; see also Table 6 below for an explanation of the abbreviations used in the Figure

Figure 7: Overview of the constructs for expression of VLPs. DETAILED DESCRIPTION OF THE INVENTION

For the purpose of vaccinating a subject against a pathogen, specific antigenic material derived from said pathogen is presented to the immune system of a subject during vaccination. As a result of this stimulation, the immune system of the subject generates an immune response in the course of which adaptive immunity to the pathogen from which the specific antigenic material is derived is developed. This principle is well known in the art and has been successfully employed in prevention and control of infectious diseases such as smallpox, polio, measles and tetanus. However, adaptive immunity is usually limited to the specific antigenic material comprised in the vaccine composition. This severely affects the long-term benefit of vaccines against pathogens that exhibit a high degree of antigenic diversity. For example, such antigenic diversity can be manifested by simultaneous circulation of different strains of the pathogen and/or by frequent occurrence of minor and/or major changes in the structure of relevant antigens. Especially in the case of influenza virus, adaptive immunity resulting from vaccination is mostly limited to those influenza strains that have been represented in the vaccine by including in the vaccine viral antigens derived from these strains. Consequently, current influenza vaccines confer only„narrow" immunity against influenza infections caused by a small number of influenza strains. The present inventors found that the vaccine composition according to the present invention confers immunity against influenza infection which is „broader" than protection conferred by current influenza vaccines. In other words, the vaccine composition according to the present invention is useful in generating immunity both against influenza strains that are represented in the vaccine composition and against influenza strains that are not represented in the vaccine composition. This is achieved by specifically selecting variants of a subtype of an influenza virus antigen according to the explanations set out in the present application, and combining these variants in a vaccine composition.

The inventors of the present application succeeded in providing a novel influenza vaccine composition comprising

(a) at least three variants of a subtype of an influenza A virus antigen, wherein the at least three variants are chosen according to the following rule:

(i) a first and second variant of a subtype of an influenza A virus antigen differ from each other by at least 10 amino acids over their entire length;

(ii) a second and third variant of a subtype of an influenza A virus antigen differ from each other by at least 10 amino acids over their entire length; and (iii) a first and third variant of a subtype of an influenza A virus antigen differ from each other by more than 35 amino acids over their entire length;

and/or

(b) at least three variants of a subtype of an influenza B virus antigen, wherein

(i) a first and second variant of a subtype of an influenza B virus antigen differ from each other by at least 10 amino acids over their entire length;

(ii) a second and third variant of a subtype of an influenza B virus antigen differ from each other by at least 10 amino acids over their entire length; and

(iii) a first and third variant of a subtype of an influenza B virus antigen differ from each other by more than 20 amino acids over their entire length.

Influenza virus is a member of the orthomyxoviridae family. There are three subtypes of influenza viruses, designated influenza A, influenza B, and influenza C. The present invention focuses on influenza A and/or influenza B virus. Thus, when reference is made herein, e.g. to influenza virus, influenza virus vaccine or the like, preferably influenza A virus and/or influenza B virus is meant. The skilled person will understand from the context in which influenza A virus or influenza B virus, respectively, is used whether influenza A virus, influenza B virus, or both are meant. The influenza virion contains a segmented negative- sense RNA genome, which encodes the following proteins: hemagglutinin (HA), neuraminidase (NA), matrix (Ml), proton ion- channel protein (M2), nucleoprotein (NP), polymerase basic protein 1 (PB 1 ), polymerase basic protein 2 (PB2), polymerase acidic protein (PA), and nonstructural protein 2 (NS2). The HA, NA, Ml, and M2 are membrane associated, whereas NP, PB1 , PB2, PA, and NS2 are nucleocapsid associated proteins. The Ml protein is the most abundant protein in influenza particles. The HA and NA proteins are envelope glycoproteins, responsible for virus attachment and penetration of the viral particles into the cell, and the sources of the major immunodominant epitopes for virus neutralization and protective immunity. Both HA and NA proteins are considered the most important components for prophylactic influenza vaccines.

Influenza A viruses infect a wide variety of subjects including fowls and mammals, including, but not limited to humans, horses, marine mammals, pigs, ferrets, and chicken, ducks, birds, gooses, etc. In animals, most influenza A viruses cause mild localized infections of the respiratory and intestinal tract. However, highly pathogenic influenza A strains, such as H5N1 , cause systemic infections in poultry in which mortality may reach 100%. Animals infected with influenza A often act as a reservoir for the influenza viruses and certain subtypes have been shown to cross the species barrier to humans.

An "influenza virus antigen" when used herein refers to a protein from influenza virus which elicits an immune response by a subject as mentioned herein. The immune response may be a cellular immune response or a humoral immune response or both. In the context of the present invention, it may rather be a humoral immune response. In addition to the surface proteins HA and NA, influenza virus comprises six additional internal genes, which give rise to eight different proteins, including polymerase genes PB1 , PB2 and PA, matrix proteins M1 and M2, nucleoprotein (NP), and non- structural proteins NS1 and NS2 (Horimoto et al., Clin Microbiol Rev. 14(1 ): 129-149, 2001 ). Each of HA, NA, PB1 , PB2, PA, M1 , M2, NP may be an influenza virus antigen. Preferably, an influenza virus antigen is a surface glycoprotein, which is preferably, hemagglutinin (HA) and/or neuraminidase (NA). Influenza A viruses can be classified into subtypes based on allelic variations in antigenic regions of two genes that encode surface glycoproteins, namely, hemagglutinin (HA) and neuraminidase (NA) which are required for viral attachment and cellular release. Currently, eighteen subtypes of HA (H1-H18) according to the WHO and eleven NA (N1-N1 1 ) subtypes are known for influenza A virus. Thus, when used herein, the term "subtype" or "subtype of an influenza virus antigen" refers to allelic variations in antigenic regions of two genes that encode surface glycoproteins, namely, hemagglutinin (HA) and neuraminidase (NA).

As mentioned herein, by "influenza virus" preferably influenza A virus or influenza B virus, respectively, is meant. The skilled person will understand from the context, when "influenza virus" is used, whether influenza A virus, influenza B virus or both are meant.

Of note, although influenza B virus is usually not classified into subtypes, but lineages, the term "subtype" is also used for influenza B virus hemagglutinin (HA) and neuraminidase (NA).

HA is a viral surface glycoprotein generally comprising on average approximately 560 amino acids and representing 25% of the total virus protein. It is responsible for adhesion of the viral particle to, and its penetration into, a host cell in the early stages of infection.

Neuraminidase (NA) is a second membrane glycoprotein of the influenza viruses. For most influenza A viruses, NA is 413 amino acid in length, and is encoded by a gene of 1413 nucleotides. NA is involved in the destruction of the cellular receptor for the viral HA by cleaving terminal neuraminic acid (also called sialic acid) residues from carbohydrate moieties on the surfaces of infected cells. NA also cleaves sialic acid residues from viral proteins, preventing aggregation of viruses.

Any subtype of an influenza virus antigen can be used for providing a vaccine composition according to the present invention. If the influenza virus antigen is HA, the subtype can be any one of H1 , H2, H3, H4, H5, H6, H7, H8, H9, H10, H1 1 , H12, H13, H14, H15, H16, H17 or H18 according to WHO. In a preferred embodiment, the HA subtype is H1 or H3 or B.

If the influenza virus antigen is NA, the subtype can be any one of NA1 , NA2, NA3, NA4, NA5, NA6, NA7, NA8, NA9, NA10, or NA1 1 according to WHO. In a preferred embodiment, the NA subtype is NA1.

As used herein, the term ..variant" or ..variant of a subtype of an influenza virus antigen" refers to an amino acid sequence variant of a subtype of an influenza virus antigen. In other words, variants differ from each other by at least one amino acid over their entire length. Such variants may result from any known mechanism. For example, the variants may naturally occur, i.e. they may be associated with circulating influenza strains. It is also possible that the variants are newly generated, e.g., by antigenic drift, antigenic shift, antigenic diversity or by genetic manipulation. As such, the variants can be the result of targeted genetic manipulation or random genetic manipulation. An "influenza virus strain " or, as also used herein in the context of influenza virus the term "strain" refers to an influenza virus that is characterized by its HA and NA subytoe, e.g. H1 N1 or H3N2.

In a preferred embodiment, if naturally occurred variants are chosen, the first variant is more ancient than the second and third variant and the second variant is more ancient than the third variant. In other words, the first variant has been circulating before the second and third variant, and the second variant has been circulating before the third variant. Thus, variants designated by a higher number are preferably more modern (in terms of time) than variants designated by a lower number. Consequently, if the vaccine composition comprises more than three variants, any additional variants, e.g., a fourth, fifth, sixth, seventh, eighth, ninth or tenth variant, are preferably more modern than any variant designated by a lower number.

In a preferred embodiment, if artificially designed variants are chosen, the same rules for amino acid differences as described herein apply. For example, if a first variant is chosen, the second variant will have the amino acid differences to the first variant as described herein. Similarly, the first variant will have the amino acid differences to the third variant as described herein, etc. A vaccine composition may further include a pharmaceutically acceptable carrier. Pharmaceutically acceptable carriers include, but are not limited to, saline, buffered saline, dextrose, water, glycerol, ethanol, and combinations thereof. The carrier and composition can be sterile, and the formulation suits the mode of administration. The composition can also contain minor amounts of wetting or emulsifying agents, or pH buffering agents. The composition can be a liquid solution, suspension, emulsion, tablet, pill, capsule, sustained release formulation, or powder. The composition can be formulated as a suppository, with traditional binders and carriers such as triglycerides. Oral formulations can include standard carriers such as pharmaceutical grades of mannitol, lactose, starch, magnesium stearate, sodium saccharine, cellulose, and magnesium.

First, second and third variant of a subtype of an influenza A virus antigen In a preferred embodiment the at least three variants of a subtype of an influenza A virus antigen are chosen according to the following rule:

(i) a first and second variant of a subtype of an influenza A virus antigen differ from each other by at least 12 amino acids over their entire length;

(ii) a second and third variant of a subtype of an influenza A virus antigen differ from each other by at least 12 amino acids over their entire length; and

(iii) a first and third variant of a subtype of an influenza A virus antigen differ from each other by more than 35 amino acids over their entire length.

In a preferred embodiment the at least three variants of a subtype of an influenza A virus antigen are chosen according to the following rule:

(i) a first and second variant of a subtype of an influenza A virus antigen differ from each other by at least 1 amino acids over their entire length;

(ii) a second and third variant of a subtype of an influenza A virus antigen differ from each other by at least 1 amino acids over their entire length; and

(iii) a first and third variant of a subtype of an influenza A virus antigen differ from each other by more than 35 amino acids over their entire length.

In a preferred embodiment the at least three variants of a subtype of an influenza A virus antigen are chosen according to the following rule:

(i) a first and second variant of a subtype of an influenza A virus antigen differ from each other by at least 16 amino acids over their entire length;

(ii) a second and third variant of a subtype of an influenza A virus antigen differ from each other by at least 16 amino acids over their entire length; and (iii) a first and third variant of a subtype of an influenza A virus antigen differ from each other by more than 35 amino acids over their entire length.

In a preferred embodiment the at least three variants of a subtype of an influenza A virus antigen are chosen according to the following rule:

(i) a first and second variant of a subtype of an influenza A virus antigen differ from each other by at least 18 amino acids over their entire length;

(ii) a second and third variant of a subtype of an influenza A virus antigen differ from each other by at least 18 amino acids over their entire length; and

(iii) a first and third variant of a subtype of an influenza A virus antigen differ from each other by more than 35 amino acids over their entire length.

In a preferred embodiment the at least three variants of a subtype of an influenza A virus antigen are chosen according to the following rule:

(i) a first and second variant of a subtype of an influenza A virus antigen differ from each other by at least 20 amino acids over their entire length;

(ii) a second and third variant of a subtype of an influenza A virus antigen differ from each other by at least 20 amino acids over their entire length; and

(iii) a first and third variant of a subtype of an influenza A virus antigen differ from each other by more than 35 amino acids over their entire length.

In a preferred embodiment the at least three variants of a subtype of an influenza A virus antigen are chosen according to the following rule:

(i) a first and second variant of a subtype of an influenza A virus antigen differ from each other by at least 10 amino acids over their entire length;

(ii) a second and third variant of a subtype of an influenza A virus antigen differ from each other by at least 10 amino acids over their entire length; and

(iii) a first and third variant of a subtype of an influenza A virus antigen differ from each other by more than 40 amino acids over their entire length.

In a preferred embodiment the at least three variants of a subtype of an influenza A virus antigen are chosen according to the following rule:

(i) a first and second variant of a subtype of an influenza A virus antigen differ from each other by at least 12 amino acids over their entire length;

(ii) a second and third variant of a subtype of an influenza A virus antigen differ from each other by at least 12 amino acids over their entire length; and

(iii) a first and third variant of a subtype of an influenza A virus antigen differ from each other by more than 40 amino acids over their entire length. In a preferred embodiment the at least three variants of a subtype of an influenza A virus antigen are chosen according to the following rule:

(i) a first and second variant of a subtype of an influenza A virus antigen differ from each other by at least 14 amino acids over their entire length;

(ii) a second and third variant of a subtype of an influenza A virus antigen differ from each other by at least 14 amino acids over their entire length; and

(iii) a first and third variant of a subtype of an influenza A virus antigen differ from each other by more than 40 amino acids over their entire length. In a preferred embodiment the at least three variants of a subtype of an influenza A virus antigen are chosen according to the following rule:

(i) a first and second variant of a subtype of an influenza A virus antigen differ from each other by at least 16 amino acids over their entire length;

(ii) a second and third variant of a subtype of an influenza A virus antigen differ from each other by at least 16 amino acids over their entire length; and

(iii) a first and third variant of a subtype of an influenza A virus antigen differ from each other by more than 40 amino acids over their entire length.

In a preferred embodiment the at least three variants of a subtype of an influenza A virus antigen are chosen according to the following rule:

(i) a first and second variant of a subtype of an influenza A virus antigen differ from each other by at least 18 amino acids over their entire length;

(ii) a second and third variant of a subtype of an influenza A virus antigen differ from each other by at least 18 amino acids over their entire length; and

(iii) a first and third variant of a subtype of an influenza A virus antigen differ from each other by more than 40 amino acids over their entire length.

In a preferred embodiment the at least three variants of a subtype of an influenza A virus antigen are chosen according to the following rule:

(i) a first and second variant of a subtype of an influenza A virus antigen differ from each other by at least 20 amino acids over their entire length;

(ii) a second and third variant of a subtype of an influenza A virus antigen differ from each other by at least 20 amino acids over their entire length; and

(iii) a first and third variant of a subtype of an influenza A virus antigen differ from each other by more than 40 amino acids over their entire length.

In a preferred embodiment the at least three variants of a subtype of an influenza A virus antigen are chosen according to the following rule: (i) a first and second variant of a subtype of an influenza A virus antigen differ from each other by at least 10 amino acids over their entire length;

(ii) a second and third variant of a subtype of an influenza A virus antigen differ from each other by at least 10 amino acids over their entire length; and

(iii) a first and third variant of a subtype of an influenza A virus antigen differ from each other by more than 45 amino acids over their entire length.

In a preferred embodiment the at least three variants of a subtype of an influenza A virus antigen are chosen according to the following rule:

(i) a first and second variant of a subtype of an influenza A virus antigen differ from each other by at least 12 amino acids over their entire length;

(ii) a second and third variant of a subtype of an influenza A virus antigen differ from each other by at least 12 amino acids over their entire length; and

(iii) a first and third variant of a subtype of an influenza A virus antigen differ from each other by more than 45 amino acids over their entire length.

In a preferred embodiment the at least three variants of a subtype of an influenza A virus antigen are chosen according to the following rule:

(i) a first and second variant of a subtype of an influenza A virus antigen differ from each other by at least 14 amino acids over their entire length;

(ii) a second and third variant of a subtype of an influenza A virus antigen differ from each other by at least 14 amino acids over their entire length; and

(iii) a first and third variant of a subtype of an influenza A virus antigen differ from each other by more than 45 amino acids over their entire length.

In a preferred embodiment the at least three variants of a subtype of an influenza A virus antigen are chosen according to the following rule:

(i) a first and second variant of a subtype of an influenza A virus antigen differ from each other by at least 16 amino acids over their entire length;

(ii) a second and third variant of a subtype of an influenza A virus antigen differ from each other by at least 16 amino acids over their entire length; and

(iii) a first and third variant of a subtype of an influenza A virus antigen differ from each other by more than 45 amino acids over their entire length. In a preferred embodiment the at least three variants of a subtype of an influenza A virus antigen are chosen according to the following rule:

(i) a first and second variant of a subtype of an influenza A virus antigen differ from each other by at least 18 amino acids over their entire length; (ii) a second and third variant of a subtype of an influenza A virus antigen differ from each other by at least 18 amino acids over their entire length; and

(iii) a first and third variant of a subtype of an influenza A virus antigen differ from each other by more than 45 amino acids over their entire length.

In a preferred embodiment the at least three variants of a subtype of an influenza A virus antigen are chosen according to the following rule:

(i) a first and second variant of a subtype of an influenza A virus antigen differ from each other by at least 20 amino acids over their entire length;

(ii) a second and third variant of a subtype of an influenza A virus antigen differ from each other by at least 20 amino acids over their entire length; and

(iii) a first and third variant of a subtype of an influenza A virus antigen differ from each other by more than 45 amino acids over their entire length. However, preferably the first and second variant of a subtype of an influenza A virus antigen differ from each other by no more than 60, 55, 50, 45, 40, 35, or 30 amino acids over their entire length.

However, preferably the first and third variant of a subtype of an influenza A virus antigen differ from each other by no more than 70, 65, 60, 55, or 50 amino acids over their entire length.

First, third and fourth variant of a subtype of an influenza A virus antigen In a preferred embodiment, the vaccine composition according to the present invention comprises a fourth variant of a subtype of an influenza A virus antigen, wherein the fourth variant (i) differs from the third variant by at least 12 amino acids over their entire length and

(ii) differs from the first variant by at least 40 amino acids over their entire length. In a preferred embodiment, the vaccine composition according to the present invention comprises a fourth variant of a subtype of an influenza A virus antigen, wherein the fourth variant (i) differs from the third variant by at least 14 amino acids over their entire length and (ii) differs from the first variant by at least 40 amino acids over their entire length. In a preferred embodiment, the vaccine composition according to the present invention comprises a fourth variant of a subtype of an influenza A virus antigen, wherein the fourth variant (i) differs from the third variant by at least 16 amino acids over their entire length and (ii) differs from the first variant by at least 40 amino acids over their entire length. In a preferred embodiment, the vaccine composition according to the present invention comprises a fourth variant of a subtype of an influenza A virus antigen, wherein the fourth variant (i) differs from the third variant by at least 18 amino acids over their entire length and (ii) differs from the first variant by at least 40 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fourth variant of a subtype of an influenza A virus antigen, wherein the fourth variant (i) differs from the third variant by at least 20 amino acids over their entire length and (ii) differs from the first variant by at least 40 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fourth variant of a subtype of an influenza A virus antigen, wherein the fourth variant (i) differs from the third variant by at least 10 amino acids over their entire length and (ii) differs from the first variant by at least 45 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fourth variant of a subtype of an influenza A virus antigen, wherein the fourth variant (i) differs from the third variant by at least 12 amino acids over their entire length and (ii) differs from the first variant by at least 45 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fourth variant of a subtype of an influenza A virus antigen, wherein the fourth variant (i) differs from the third variant by at least 14 amino acids over their entire length and (ii) differs from the first variant by at least 45 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fourth variant of a subtype of an influenza A virus antigen, wherein the fourth variant (i) differs from the third variant by at least 16 amino acids over their entire length and (ii) differs from the first variant by at least 45 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fourth variant of a subtype of an influenza A virus antigen, wherein the fourth variant (i) differs from the third variant by at least 18 amino acids over their entire length and (ii) differs from the first variant by at least 45 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fourth variant of a subtype of an influenza A virus antigen, wherein the fourth variant (i) differs from the third variant by at least 20 amino acids over their entire length and (ii) differs from the first variant by at least 45 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fourth variant of a subtype of an influenza A virus antigen, wherein the fourth variant (i) differs from the third variant by at least 10 amino acids over their entire length and (ii) differs from the first variant by at least 50 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fourth variant of a subtype of an influenza A virus antigen, wherein the fourth variant (i) differs from the third variant by at least 12 amino acids over their entire length and (ii) differs from the first variant by at least 50 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fourth variant of a subtype of an influenza A virus antigen, wherein the fourth variant (i) differs from the third variant by at least 14 amino acids over their entire length and (ii) differs from the first variant by at least 50 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fourth variant of a subtype of an influenza A virus antigen, wherein the fourth variant (i) differs from the third variant by at least 16 amino acids over their entire length and (ii) differs from the first variant by at least 50 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fourth variant of a subtype of an influenza A virus antigen, wherein the fourth variant (i) differs from the third variant by at least 18 amino acids over their entire length and (ii) differs from the first variant by at least 50 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fourth variant of a subtype of an influenza A virus antigen, wherein the fourth variant (i) differs from the third variant by at least 20 amino acids over their entire length and (ii) differs from the first variant by at least 50 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fourth variant of a subtype of an influenza A virus antigen, wherein the fourth variant (i) differs from the third variant by at least 10 amino acids over their entire length and (ii) differs from the first variant by at least 55 amino acids over their entire length. In a preferred embodiment, the vaccine composition according to the present invention comprises a fourth variant of a subtype of an influenza A virus antigen, wherein the fourth variant (i) differs from the third variant by at least 12 amino acids over their entire length and (ii) differs from the first variant by at least 55 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fourth variant of a subtype of an influenza A virus antigen, wherein the fourth variant (i) differs from the third variant by at least 14 amino acids over their entire length and (ii) differs from the first variant by at least 55 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fourth variant of a subtype of an influenza A virus antigen, wherein the fourth variant (i) differs from the third variant by at least 16 amino acids over their entire length and (ii) differs from the first variant by at least 55 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fourth variant of a subtype of an influenza A virus antigen, wherein the fourth variant (i) differs from the third variant by at least 18 amino acids over their entire length and (ii) differs from the first variant by at least 55 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fourth variant of a subtype of an influenza A virus antigen, wherein the fourth variant (i) differs from the third variant by at least 20 amino acids over their entire length and (ii) differs from the first variant by at least 55 amino acids over their entire length.

However, preferably the third and fourth variant of a subtype of an influenza A virus antigen differ from each other by no more than 60, 55, 50, 45, 40, 35, or 30 amino acids over their entire length. However, preferably the first and fourth variant of a subtype of an influenza A virus antigen differ from each other by no more than 70, 65 or 60 amino acids over their entire length.

First, fourth and fifth variant of a subtype of an influenza A virus antigen In a preferred embodiment, the vaccine composition according to the present invention comprises a fifth variant of a subtype of an influenza A virus antigen, wherein the fifth variant (i) differs from the fourth variant by at least 1 1 amino acids over their entire length and (ii) differs from the first variant by at least 50 amino acids over their entire length. In a preferred embodiment, the vaccine composition according to the present invention comprises a fifth variant of a subtype of an influenza A virus antigen, wherein the fifth variant (i) differs from the fourth variant by at least 12 amino acids over their entire length and (ii) differs from the first variant by at least 50 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fifth variant of a subtype of an influenza A virus antigen, wherein the fifth variant (i) differs from the fourth variant by at least 15 amino acids over their entire length and (ii) differs from the first variant by at least 50 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fifth variant of a subtype of an influenza A virus antigen, wherein the fifth variant (i) differs from the fourth variant by at least 20 amino acids over their entire length and (ii) differs from the first variant by at least 50 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fifth variant of a subtype of an influenza A virus antigen, wherein the fifth variant (i) differs from the fourth variant by at least 10 amino acids over their entire length and (ii) differs from the first variant by at least 55 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fifth variant of a subtype of an influenza A virus antigen, wherein the fifth variant (i) differs from the fourth variant by at least 1 1 amino acids over their entire length and (ii) differs from the first variant by at least 55 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fifth variant of a subtype of an influenza A virus antigen, wherein the fifth variant (i) differs from the fourth variant by at least 12 amino acids over their entire length and (ii) differs from the first variant by at least 55 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fifth variant of a subtype of an influenza A virus antigen, wherein the fifth variant (i) differs from the fourth variant by at least 15 amino acids over their entire length and (ii) differs from the first variant by at least 55 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fifth variant of a subtype of an influenza A virus antigen, wherein the fifth variant (i) differs from the fourth variant by at least 20 amino acids over their entire length and (ii) differs from the first variant by at least 55 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fifth variant of a subtype of an influenza A virus antigen, wherein the fifth variant (i) differs from the fourth variant by at least 10 amino acids over their entire length and (ii) differs from the first variant by at least 60 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fifth variant of a subtype of an influenza A virus antigen, wherein the fifth variant (i) differs from the fourth variant by at least 1 1 amino acids over their entire length and (ii) differs from the first variant by at least 60 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fifth variant of a subtype of an influenza A virus antigen, wherein the fifth variant (i) differs from the fourth variant by at least 12 amino acids over their entire length and (ii) differs from the first variant by at least 60 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fifth variant of a subtype of an influenza A virus antigen, wherein the fifth variant (i) differs from the fourth variant by at least 15 amino acids over their entire length and (ii) differs from the first variant by at least 60 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fifth variant of a subtype of an influenza A virus antigen, wherein the fifth variant (i) differs from the fourth variant by at least 20 amino acids over their entire length and (ii) differs from the first variant by at least 60 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fifth variant of a subtype of an influenza A virus antigen, wherein the fifth variant (i) differs from the fourth variant by at least 10 amino acids over their entire length and (ii) differs from the first variant by at least 65 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fifth variant of a subtype of an influenza A virus antigen, wherein the fifth variant (i) differs from the fourth variant by at least 1 1 amino acids over their entire length and (ii) differs from the first variant by at least 65 amino acids over their entire length. In a preferred embodiment, the vaccine composition according to the present invention comprises a fifth variant of a subtype of an influenza A virus antigen, wherein the fifth variant (i) differs from the fourth variant by at least 12 amino acids over their entire length and (ii) differs from the first variant by at least 65 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fifth variant of a subtype of an influenza A virus antigen, wherein the fifth variant (i) differs from the fourth variant by at least 15 amino acids over their entire length and (ii) differs from the first variant by at least 65 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fifth variant of a subtype of an influenza A virus antigen, wherein the fifth variant (i) differs from the fourth variant by at least 20 amino acids over their entire length and (ii) differs from the first variant by at least 65 amino acids over their entire length.

However, preferably the fourth and fifth variant of a subtype of an influenza A virus antigen differ from each other by no more than 60, 55, 50, 45, 40, 35 or 30 amino acids over their entire length. However, preferably the first and fifth variant of a subtype of an influenza A virus antigen differ from each other by no more than 90, 85, 80, 75 or 70 amino acids over their entire length.

Sixth, seventh, eighth, ninth or tenth variant of a subtype of an influenza A virus antigen

In another preferred embodiment, the vaccine composition according to the present invention comprises a sixth, seventh, eighth, ninth or tenth variant of a subtype of an influenza A virus antigen or several of these variants. It is envisioned that the sixth variant of a subtype of an influenza A virus antigen differs from the fifth variant by at least 10 or 20 amino acids, but by no more than 50 or 40 amino acids, e.g. 10-40, 20-40, 10-50 or 20-50 amino acids difference.

It is envisioned that the seventh variant of a subtype of an influenza A virus antigen differs from the sixth variant by at least 10 or 20 amino acids, but by no more than 50 or 40 amino acids, e.g. 10-40, 20-40, 10-50 or 20-50 amino acids difference. It is envisioned that the eighth variant of a subtype of an influenza A virus antigen differs from the seventh variant by at least 10 or 20 amino acids, but by no more than 50 or 40 amino acids, e.g. 10-40, 20-40, 10-50 or 20-50 amino acids difference. It is envisioned that the ninth variant of a subtype of an influenza A virus antigen differs from the eighth variant by at least 10 or 20 amino acids, but by no more than 50 or 40 amino acids, e.g. 10-40, 20-40, 10-50 or 20-50 amino acids difference.

It is envisioned that the tenth variant of a subtype of an influenza A virus antigen differs from the ninth variant by at least 10 or 20 amino acids, but by no more than 50 or 40 amino acids, e.g. 10-40, 20-40, 10-50 or 20-50 amino acids difference.

In a preferred embodiment, the number of variants of the subtype of said influenza A virus antigen exceeds the number of variants of the subtype of said influenza virus antigens that is comprised in said vaccine.

First, second and third variant of a subtype of an influenza B virus antigen

In a preferred embodiment the at least three variants of a subtype of an influenza B virus antigen are chosen according to the following rule:

(i) a first and second variant of a subtype of an influenza B virus antigen differ from each other by at least 12 amino acids over their entire length;

(ii) a second and third variant of a subtype of an influenza B virus antigen differ from each other by at least 12 amino acids over their entire length; and

(iii) a first and third variant of a subtype of an influenza B virus antigen differ from each other by more than 20 amino acids over their entire length.

In a preferred embodiment the at least three variants of a subtype of an influenza B virus antigen are chosen according to the following rule:

(i) a first and second variant of a subtype of an influenza B virus antigen differ from each other by at least 14 amino acids over their entire length;

(ii) a second and third variant of a subtype of an influenza B virus antigen differ from each other by at least 14 amino acids over their entire length; and

(iii) a first and third variant of a subtype of an influenza B virus antigen differ from each other by more than 20 amino acids over their entire length.

In a preferred embodiment the at least three variants of a subtype of an influenza B virus antigen are chosen according to the following rule: (i) a first and second variant of a subtype of an influenza B virus antigen differ from each other by at least 16 amino acids over their entire length;

(ii) a second and third variant of a subtype of an influenza B virus antigen differ from each other by at least 16 amino acids over their entire length; and

(iii) a first and third variant of a subtype of an influenza B virus antigen differ from each other by more than 20 amino acids over their entire length.

In a preferred embodiment the at least three variants of a subtype of an influenza B virus antigen are chosen according to the following rule:

(i) a first and second variant of a subtype of an influenza B virus antigen differ from each other by at least 18 amino acids over their entire length;

(ii) a second and third variant of a subtype of an influenza B virus antigen differ from each other by at least 18 amino acids over their entire length; and

(iii) a first and third variant of a subtype of an influenza B virus antigen differ from each other by more than 20 amino acids over their entire length.

In a preferred embodiment the at least three variants of a subtype of an influenza B virus antigen are chosen according to the following rule:

(i) a first and second variant of a subtype of an influenza B virus antigen differ from each other by at least 20 amino acids over their entire length;

(ii) a second and third variant of a subtype of an influenza B virus antigen differ from each other by at least 20 amino acids over their entire length; and

(iii) a first and third variant of a subtype of an influenza B virus antigen differ from each other by more than 20 amino acids over their entire length.

In a preferred embodiment the at least three variants of a subtype of an influenza B virus antigen are chosen according to the following rule:

(i) a first and second variant of a subtype of an influenza B virus antigen differ from each other by at least 10 amino acids over their entire length;

(ii) a second and third variant of a subtype of an influenza B virus antigen differ from each other by at least 10 amino acids over their entire length; and

(iii) a first and third variant of a subtype of an influenza B virus antigen differ from each other by more than 25 amino acids over their entire length. In a preferred embodiment the at least three variants of a subtype of an influenza B virus antigen are chosen according to the following rule:

(i) a first and second variant of a subtype of an influenza B virus antigen differ from each other by at least 12 amino acids over their entire length; (ii) a second and third variant of a subtype of an influenza B virus antigen differ from each other by at least 12 amino acids over their entire length; and

(iii) a first and third variant of a subtype of an influenza B virus antigen differ from each other by more than 25 amino acids over their entire length.

In a preferred embodiment the at least three variants of a subtype of an influenza B virus antigen are chosen according to the following rule:

(i) a first and second variant of a subtype of an influenza B virus antigen differ from each other by at least 14 amino acids over their entire length;

(ii) a second and third variant of a subtype of an influenza B virus antigen differ from each other by at least 14 amino acids over their entire length; and

(iii) a first and third variant of a subtype of an influenza B virus antigen differ from each other by more than 25 amino acids over their entire length. In a preferred embodiment the at least three variants of a subtype of an influenza B virus antigen are chosen according to the following rule:

(i) a first and second variant of a subtype of an influenza B virus antigen differ from each other by at least 16 amino acids over their entire length;

(ii) a second and third variant of a subtype of an influenza B virus antigen differ from each other by at least 16 amino acids over their entire length; and

(iii) a first and third variant of a subtype of an influenza B virus antigen differ from each other by more than 25 amino acids over their entire length.

In a preferred embodiment the at least three variants of a subtype of an influenza B virus antigen are chosen according to the following rule:

(i) a first and second variant of a subtype of an influenza B virus antigen differ from each other by at least 18 amino acids over their entire length;

(ii) a second and third variant of a subtype of an influenza B virus antigen differ from each other by at least 18 amino acids over their entire length; and

(iii) a first and third variant of a subtype of an influenza B virus antigen differ from each other by more than 25 amino acids over their entire length.

In a preferred embodiment the at least three variants of a subtype of an influenza B virus antigen are chosen according to the following rule:

(i) a first and second variant of a subtype of an influenza B virus antigen differ from each other by at least 20 amino acids over their entire length;

(ii) a second and third variant of a subtype of an influenza B virus antigen differ from each other by at least 20 amino acids over their entire length; and (iii) a first and third variant of a subtype of an influenza B virus antigen differ from each other by more than 25 amino acids over their entire length.

In a preferred embodiment the at least three variants of a subtype of an influenza B virus antigen are chosen according to the following rule:

(i) a first and second variant of a subtype of an influenza B virus antigen differ from each other by at least 10 amino acids over their entire length;

(ii) a second and third variant of a subtype of an influenza B virus antigen differ from each other by at least 10 amino acids over their entire length; and

(iii) a first and third variant of a subtype of an influenza B virus antigen differ from each other by more than 30 amino acids over their entire length.

In a preferred embodiment the at least three variants of a subtype of an influenza B virus antigen are chosen according to the following rule:

(i) a first and second variant of a subtype of an influenza B virus antigen differ from each other by at least 12 amino acids over their entire length;

(ii) a second and third variant of a subtype of an influenza B virus antigen differ from each other by at least 12 amino acids over their entire length; and

(iii) a first and third variant of a subtype of an influenza B virus antigen differ from each other by more than 30 amino acids over their entire length.

In a preferred embodiment the at least three variants of a subtype of an influenza B virus antigen are chosen according to the following rule:

(i) a first and second variant of a subtype of an influenza B virus antigen differ from each other by at least 14 amino acids over their entire length;

(ii) a second and third variant of a subtype of an influenza B virus antigen differ from each other by at least 1 amino acids over their entire length; and

(iii) a first and third variant of a subtype of an influenza B virus antigen differ from each other by more than 30 amino acids over their entire length. In a preferred embodiment the at least three variants of a subtype of an influenza B virus antigen are chosen according to the following rule:

(i) a first and second variant of a subtype of an influenza B virus antigen differ from each other by at least 16 amino acids over their entire length;

(ii) a second and third variant of a subtype of an influenza B virus antigen differ from each other by at least 16 amino acids over their entire length; and

(iii) a first and third variant of a subtype of an influenza B virus antigen differ from each other by more than 30 amino acids over their entire length. In a preferred embodiment the at least three variants of a subtype of an influenza B virus antigen are chosen according to the following rule:

(i) a first and second variant of a subtype of an influenza B virus antigen differ from each other by at least 18 amino acids over their entire length;

(ii) a second and third variant of a subtype of an influenza B virus antigen differ from each other by at least 18 amino acids over their entire length; and

(iii) a first and third variant of a subtype of an influenza B virus antigen differ from each other by more than 30 amino acids over their entire length. In a preferred embodiment the at least three variants of a subtype of an influenza B virus antigen are chosen according to the following rule:

(i) a first and second variant of a subtype of an influenza B virus antigen differ from each other by at least 20 amino acids over their entire length;

(ii) a second and third variant of a subtype of an influenza B virus antigen differ from each other by at least 20 amino acids over their entire length; and

(iii) a first and third variant of a subtype of an influenza B virus antigen differ from each other by more than 30 amino acids over their entire length.

However, preferably the first and second variant of a subtype of an influenza B virus antigen differ from each other by no more than 60, 55, 50, 45, 40, 35, or 30 amino acids over their entire length.

However, preferably the first and third variant of a subtype of an influenza B virus antigen differ from each other by no more than 70, 65, 60, 55, 50, 45, 40 or 35 amino acids over their entire length.

First, third and fourth variant of a subtype of an influenza B virus antigen

In a preferred embodiment, the vaccine composition according to the present invention comprises a fourth variant of a subtype of an influenza B virus antigen, wherein the fourth variant (i) differs from the third variant by at least 12 amino acids over their entire length and (ii) differs from the first variant by at least 20 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fourth variant of a subtype of an influenza B virus antigen, wherein the fourth variant (i) differs from the third variant by at least 14 amino acids over their entire length and (ii) differs from the first variant by at least 20 amino acids over their entire length. In a preferred embodiment, the vaccine composition according to the present invention comprises a fourth variant of a subtype of an influenza B virus antigen, wherein the fourth variant (i) differs from the third variant by at least 16 amino acids over their entire length and (ii) differs from the first variant by at least 20 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fourth variant of a subtype of an influenza B virus antigen, wherein the fourth variant (i) differs from the third variant by at least 18 amino acids over their entire length and (ii) differs from the first variant by at least 20 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fourth variant of a subtype of an influenza B virus antigen, wherein the fourth variant (i) differs from the third variant by at least 20 amino acids over their entire length and (ii) differs from the first variant by at least 20 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fourth variant of a subtype of an influenza B virus antigen, wherein the fourth variant (i) differs from the third variant by at least 10 amino acids over their entire length and (ii) differs from the first variant by at least 25 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fourth variant of a subtype of an influenza B virus antigen, wherein the fourth variant (i) differs from the third variant by at least 12 amino acids over their entire length and (ii) differs from the first variant by at least 25 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fourth variant of a subtype of an influenza B virus antigen, wherein the fourth variant (i) differs from the third variant by at least 1 amino acids over their entire length and (ii) differs from the first variant by at least 25 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fourth variant of a subtype of an influenza B virus antigen, wherein the fourth variant (i) differs from the third variant by at least 16 amino acids over their entire length and (ii) differs from the first variant by at least 25 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fourth variant of a subtype of an influenza B virus antigen, wherein the fourth variant (i) differs from the third variant by at least 18 amino acids over their entire length and (ii) differs from the first variant by at least 25 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fourth variant of a subtype of an influenza B virus antigen, wherein the fourth variant (i) differs from the third variant by at least 20 amino acids over their entire length and (ii) differs from the first variant by at least 25 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fourth variant of a subtype of an influenza B virus antigen, wherein the fourth variant (i) differs from the third variant by at least 10 amino acids over their entire length and (ii) differs from the first variant by at least 30 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fourth variant of a subtype of an influenza B virus antigen, wherein the fourth variant (i) differs from the third variant by at least 12 amino acids over their entire length and (ii) differs from the first variant by at least 30 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fourth variant of a subtype of an influenza B virus antigen, wherein the fourth variant (i) differs from the third variant by at least 14 amino acids over their entire length and (ii) differs from the first variant by at least 30 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fourth variant of a subtype of an influenza B virus antigen, wherein the fourth variant (i) differs from the third variant by at least 16 amino acids over their entire length and (ii) differs from the first variant by at least 30 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fourth variant of a subtype of an influenza B virus antigen, wherein the fourth variant (i) differs from the third variant by at least 18 amino acids over their entire length and (ii) differs from the first variant by at least 30 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fourth variant of a subtype of an influenza B virus antigen, wherein the fourth variant (i) differs from the third variant by at least 20 amino acids over their entire length and (ii) differs from the first variant by at least 30 amino acids over their entire length. In a preferred embodiment, the vaccine composition according to the present invention comprises a fourth variant of a subtype of an influenza B virus antigen, wherein the fourth variant (i) differs from the third variant by at least 10 amino acids over their entire length and (ii) differs from the first variant by at least 35 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fourth variant of a subtype of an influenza B virus antigen, wherein the fourth variant (i) differs from the third variant by at least 12 amino acids over their entire length and (ii) differs from the first variant by at least 35 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fourth variant of a subtype of an influenza B virus antigen, wherein the fourth variant (i) differs from the third variant by at least 14 amino acids over their entire length and (ii) differs from the first variant by at least 35 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fourth variant of a subtype of an influenza B virus antigen, wherein the fourth variant (i) differs from the third variant by at least 16 amino acids over their entire length and (ii) differs from the first variant by at least 35 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fourth variant of a subtype of an influenza B virus antigen, wherein the fourth variant (i) differs from the third variant by at least 18 amino acids over their entire length and (ii) differs from the first variant by at least 35 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fourth variant of a subtype of an influenza B virus antigen, wherein the fourth variant (i) differs from the third variant by at least 20 amino acids over their entire length and (ii) differs from the first variant by at least 35 amino acids over their entire length.

However, preferably the third and fourth variant of a subtype of an influenza B virus antigen differ from each other by no more than 60, 55, 50, 45, 40, 35, or 30 amino acids over their entire length. However, preferably the first and fourth variant of a subtype of an influenza B virus antigen differ from each other by no more than 70, 65 or 60 amino acids over their entire length. First, fourth and fifth variant of a subtype of an influenza B virus antigen

In a preferred embodiment, the vaccine composition according to the present invention comprises a fifth variant of a subtype of an influenza B virus antigen, wherein the fifth variant (i) differs from the fourth variant by at least 12 amino acids over their entire length and (ii) differs from the first variant by at least 30 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fifth variant of a subtype of an influenza B virus antigen, wherein the fifth variant (i) differs from the fourth variant by at least 15 amino acids over their entire length and (ii) differs from the first variant by at least 30 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fifth variant of a subtype of an influenza B virus antigen, wherein the fifth variant (i) differs from the fourth variant by at least 20 amino acids over their entire length and (ii) differs from the first variant by at least 30 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fifth variant of a subtype of an influenza B virus antigen, wherein the fifth variant (i) differs from the fourth variant by at least 10 amino acids over their entire length and (ii) differs from the first variant by at least 35 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fifth variant of a subtype of an influenza B virus antigen, wherein the fifth variant (i) differs from the fourth variant by at least 12 amino acids over their entire length and (ii) differs from the first variant by at least 35 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fifth variant of a subtype of an influenza B virus antigen, wherein the fifth variant (i) differs from the fourth variant by at least 15 amino acids over their entire length and (ii) differs from the first variant by at least 35 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fifth variant of a subtype of an influenza B virus antigen, wherein the fifth variant (i) differs from the fourth variant by at least 20 amino acids over their entire length and (ii) differs from the first variant by at least 35 amino acids over their entire length. In a preferred embodiment, the vaccine composition according to the present invention comprises a fifth variant of a subtype of an influenza B virus antigen, wherein the fifth variant (i) differs from the fourth variant by at least 10 amino acids over their entire length and (ii) differs from the first variant by at least 40 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fifth variant of a subtype of an influenza B virus antigen, wherein the fifth variant (i) differs from the fourth variant by at least 12 amino acids over their entire length and (ii) differs from the first variant by at least 40 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fifth variant of a subtype of an influenza B virus antigen, wherein the fifth variant (i) differs from the fourth variant by at least 15 amino acids over their entire length and (ii) differs from the first variant by at least 40 amino acids over their entire length.

In a preferred embodiment, the vaccine composition according to the present invention comprises a fifth variant of a subtype of an influenza B virus antigen, wherein the fifth variant (i) differs from the fourth variant by at least 20 amino acids over their entire length and (ii) differs from the first variant by at least 40 amino acids over their entire length.

However, preferably the fourth and fifth variant of a subtype of an influenza B virus antigen differ from each other by no more than 60, 55, 50, 45, 40, 35 or 30 amino acids over their entire length. However, preferably the first and fifth variant of a subtype of an influenza B virus antigen differ from each other by no more than 100, 90, 85, 80, 75, 70, 65, 55 or 50 amino acids over their entire length.

Sixth, seventh, eighth, ninth or tenth variant of a subtype of an influenza B virus antigen

In another preferred embodiment, the vaccine composition according to the present invention comprises a sixth, seventh, eighth, ninth or tenth variant of a subtype of an influenza B virus antigen or several of these variants.

It is envisioned that the sixth variant of a subtype of an influenza B virus antigen differs from the fifth variant by at least 10 or 20 amino acids, but by no more than 50 or 40 amino acids, e.g. 10-40, 20-40, 10-50 or 20-50 amino acids difference. It is envisioned that the seventh variant of a subtype of an influenza B virus antigen differs from the sixth variant by at least 10 or 20 amino acids, but by no more than 50 or 40 amino acids, e.g. 10-40, 20-40, 10-50 or 20-50 amino acids difference.

It is envisioned that the eighth variant of a subtype of an influenza B virus antigen differs from the seventh variant by at least 10 or 20 amino acids, but by no more than 50 or 40 amino acids, e.g. 10-40, 20-40, 10-50 or 20-50 amino acids difference. It is envisioned that the ninth variant of a subtype of an influenza B virus antigen differs from the eighth variant by at least 10 or 20 amino acids, but by no more than 50 or 40 amino acids, e.g. 10-40, 20-40, 10-50 or 20-50 amino acids difference.

It is envisioned that the tenth variant of a subtype of an influenza B virus antigen differs from the ninth variant by at least 10 or 20 amino acids, but by no more than 50 or 40 amino acids, e.g. 10-40, 20-40, 10-50 or 20-50 amino acids difference.

In a preferred embodiment, the number of variants of the subtype of said influenza B virus antigen exceeds the number of variants of the subtype of said influenza virus antigens that is comprised in said vaccine.

In an especially preferred embodiment, the subtype is H1 and the vaccine composition comprises the following five variants of H1 HA:

A/Puerto Rico/8/1934 (SEQ ID No: 1 ),

A/USSR/92/1977 (SEQ ID No: 2),

A/Texas/36/1991 (SEQ ID No: 3),

A/New Caledonia/20/1999 (SEQ ID No: 4), and

A/Brisbane/59/2007 (SEQ ID No: 5). In another especially preferred embodiment, the subtype is H3 and the vaccine composition comprises the following five variants of H3 HA:

A/Hong Kong/1/1968 (SEQ ID No: 9),

A/England/321/1977 (SEQ ID No: 10),

A/Sichuan/2/1987 (SEQ ID No: 1 1 ),

A/Nanchang/933/1995 (SEQ ID No: 12) or A Johannesburg/33/94 (SEQ ID No: 13), and A/Wyoming/3/2003 (SEQ ID NO: 14) or A/Fujian/41 1/2002 (SEQ ID No: 15). In yet another especially preferred embodiment, the subtype is B and the vaccine composition comprises the following five variants of B HA:

B/Hong Kong/8/73 (SEQ ID No: 21 ),

B/Beijing/1/1987 (SEQ ID No: 22) or B/Victori a/02/1987 (SEQ ID No: 23),

B/Yamagata/16/1988 (SEQ ID No: 24),

B/Jiangsu/10/2003 (SEQ ID No: 25), and

B/Malaysia/2506/2004 (SEQ ID No: 26).

In a still especially preferred embodiment, the subtype is NA and the vaccine composition comprises the following three variants of N1 NA:

A/Puerto Rico/8/1934 (SEQ ID No: 31 ),

A/Texas/36/1991 (SEQ ID No: 32), and/or

A/Brisbane/59/2007 (SEQ ID No: 33). It is also envisioned that the vaccine composition may comprise one or more of these N1 variants.

In a yet further especially preferred embodiment, the subtype is H4 and the vaccine composition comprises the following five variants of HA:

H4: A/Mallard/Alberta/47/1991 (SEQ ID No: 34),

H7: A/cinnamon teal/Bolivia/4537/2001 (SEQ ID No: 35),

H10: A/quail/N J/25254-22/1995 (SEQ ID No: 36),

H14: A/herring gull/Astrakhan/267/1982 (SEQ ID No: 37), and/or

H15: A/Australian shelduck/Western Australia/1762/1979 (SEQ ID No: 38). It is also envisioned that the vaccine composition may comprise one or more of these H4, H7, H10, H14, H15 variants.

Sequence alignments Differences between amino acid sequences of variants as referred to herein may be translated into a certain degree of "identity". By "% identity" is meant a property of sequences that measures their similarity or relationship. Identity is measured by dividing the number of identical residues by the total number of residues and gaps and multiplying the product by 100. Preferably, identity is determined over the entire length of the sequences being compared. "Gaps" are spaces in an alignment that are the result of additions or deletions of amino acids. Thus, two copies of exactly the same sequence have 100% identity, but sequences that are less highly conserved, and have deletions, additions, or replacements, may have a lower degree of identity. Those skilled in the art will recognize that several computer programs are available for determining sequence identity using standard parameters, for example Blast (Altschul, et al. (1997) Nucleic Acids Res. 25:3389- 3402), Blast2 (Altschul, et al. (1990) J. Mol. Biol. 215:403-410), MAFFT, ClustalW or Smith- Waterman (Smith, et al. (1981 ) J. Mol. Biol. 147:195-197) or any other suitable program which is suitable to generate sequence alignments. These computer programs may be used to compare a plurality of amino acid sequences, e.g. 2 or more amino acid sequences. However, an alignment may also be done manually.

The preferred computer program for determining sequence identity is MUSCLE (multiple sequence alignment with high accuracy and high throughput); Edgar RC (2004) Nucleic Acids Res. 32: 1792-1797. The MUSCLE program is, e.g. available at www.fludb.org.

When used herein the term "over their entire length" means that amino acid sequences of at least two variants of a subtype of influenza virus antigens, except for a signal sequence or pro-sequence that may be contained in the amino acid sequence of a variant, are aligned and inspected in order to determine the degree of identity as mentioned above. For example, it is known that H1 contains a 17 amino acid pro-sequence, H3 a 16 amino acid pro-sequence and HB a 15 amino acid pro-sequence. The skilled person is aware of pro- sequences from HA and is readily in a position to determine the same in an HA polypeptide (see, e.g., Kovacova et al. (2002) Virus Genes 24:1 , 57-63)

Cross-neutralization

In a preferred embodiment, the vaccine composition according to the present invention effects cross-neutralization against at least one variant of said subtype of said influenza virus A and/or B antigen that is different from said variants comprised in said vaccine.

Cross-neutralization means that serum obtained from a subject after immunization with a vaccine composition of the present invention is capable of neutralizing at least 25% infection, more preferably at least 50% infection of MDCK cells, as determined in an influenza microneutralization assay, caused by an influenza virus having a variant of a subtype of an influenza virus antigen which is different from a variants comprised in a vaccine composition of the present invention.

A micro neutralization assay is preferably performed as follows:

Firstly, virus stocks are titrated. A series of log 10 dilutions are made and 0.1 ml/well (10 wells per dilution) are added into flat-bottomed 96-well plates containing a monolayer of confluent MDCK cells. Plates are incubated at room temperature for 30 minutes before replacing inoculum with infection medium (DMEM containing 2mM glutamine, sodium bicarbonate , penicillin-streptomycin 1/100, amphotericin B and 0.0025μg/ml TPCK trypsin). Plates are further incubated for 72 hours at 35°C. 50μΙ per well supernatants are harvested and run in HA assays using 0.7% turkey red blood cells. 50 % Tissue culture infectious doses (TCID ₅₀ ) are calculated using the Spearman-Karber formula.

Sera samples are heat treated at 56°C for 50 minutes then added in duplicate into flat- bottomed 96-well plates using a starting dilution of 1/20, followed by a further seven doubling dilutions. 10 ² TCID ₅₀ (100 μΙ) viruses is then added into each well. Plates are incubated at room temperature for 1 hour before adding the mixtures to flat-bottomed 96-well plates containing a monolayer of confluent MDCK cells. After 30 minutes incubation at room temperature the serum-virus mixture is replaced with 100μΙ of infection media and incubated for 72 hours at 35°C. Supernatants are screened using an HA assay, as above. Serum neutralisation titres are expressed as the reciprocal of the highest dilution whereby 50% infection was prevented. Titres are the average of duplicate samples. Each assay run includes a back-titration of the viruses used and validation criteria of 10 ² +/- 10° ⁵ /100μΙ.

In an even more preferred embodiment, the variant against which cross-neutralization in an influenza microneutralization assay is effected differs from each variant comprised in the vaccine composition by at least 20, 25, 30, 35, 40, 45, 50, 55, 60, 65 or 70 or more amino acids over their entire length.

Preferably, the variant of a subtype of an influenza A virus antigen against which cross- neutralization in an influenza microneutralization assay is effected differs from each variant of a subtype of an influenza A virus antigen comprised in the vaccine composition by at least 20 amino acids over their entire length. Said variant of a subtype of an influenza A virus antigen against which cross-neutralization in an influenza microneutralization assay is effected may also differ from each variant of a subtype of an influenza A virus antigen comprised in the vaccine composition by at least 25, 30, 40, 50, 60, 70, 80, 90 or even 100 amino acids. However, it may not differ by more than 120 or preferably 1 10 amino acids.

Preferably, the variant of a subtype of an influenza B virus antigen against which cross- neutralization in an influenza microneutralization assay is effected differs from each variant of a subtype of an influenza B virus antigen comprised in the vaccine composition by at least 15 amino acids over their entire length. Said variant of a subtype of an influenza B virus antigen against which cross-neutralization in an influenza microneutralization assay is effected may also differ from each variant of a subtype of an influenza A virus antigen comprised in the vaccine composition by at least 16, 18, 20, 25, 30, 35, 40 or even 50 amino acids. However, it may not differ by more than 70 or preferably 60 amino acids. The vaccine composition according to the present invention can provide vaccinated subjects with broad protection against influenza strains both circulating at the time of vaccination and predicted to arise in the future. Eventually, the vaccine of the present invention is envisaged to be useful in providing a "universal vaccine" both for seasonal and pandemic flu. Thereby, the risk of devastating pandemic outbreaks such as caused by the Spanish Flu in 1918 could be decreased. Taken together, the vaccine composition according to the present invention provides a significant improvement of public health. But also from an economic point of view, the vaccine composition according to the present invention is highly advantageous. Annual reformulation of influenza vaccines as required for control of seasonal flu is a highly cost-intensive process. Also the cost of annual distribution and administration to subjects is significant. Due to the fact that the vaccine composition according to the present invention provides vaccinated subjects with broad protection as described hereinbefore, it is no longer necessary to annually reformulate and re-administer the vaccine while still maintaining or even improving public health.

The influenza virus antigen can be comprised in the vaccine composition in any form such as in the form of one or more proteins or in the form of one or more nucleic acids or combinations of both. It is also possible that the proteins or nucleic acids or combinations of both are present in the form of aggregates. If nucleic acids are used, the rules regarding the necessary amino acid differences between the corresponding variants can be applied in an analogous manner, in consideration of basic genetic principles.

In a preferred embodiment, the variants comprised in the vaccine composition are proteins. These proteins can be produced or isolated by any method know in the art. These proteins can be natural proteins or they can be artificial proteins such as genetically engineered proteins. In a specific embodiment, the variants are totally or partially comprised in a fusion protein. To create a fusion protein, the nucleic acid sequences must be in the same reading frame. For example, a fusion protein includes an influenza HA variant or NA variant fused to a heterologous protein.

In a preferred embodiment, the variants comprised in the vaccine composition are totally or partially immobilized on the surface of a virus-like particle (VLP). A single VLP can comprise identical or different variants of a subtype of an influenza virus antigen. Thus, in a preferred embodiment, each variant comprised in the vaccine composition can be present on a distinct population of VLPs (monovalent VLP), i.e. there are several VLP populations, each representing an individual variant. In another preferred embodiment, several variants such as two, three, four, five, six, seven, eight, nine, ten or more variants can be present on the same VLP (polyvalent VLP), i.e. there is less than one VLP population per variant. For example, a specific VLP population can carry two, three, four, five, six, seven, eight, nine, ten or more variants on its surface, with five (pentavalent) being preferred. However, it is also envisioned that a vaccine composition may comprise monovalent and polyvalent VLPs that have variants as described herein totally or partially immobilized on their surface.

Virus-like particles (VLPs) and their generation are well known. They are especially useful in the field of vaccination since they can be tailored to various applications. Furthermore, they do preferably not contain any viral genetic material, preventing their replication in host cells and rendering them non-infectious, which is preferred from a medical point of view in order to reduce the risk associated with vaccinations. The main components of VLPs are viral structural proteins.

VLPs can often be produced by heterologous expression and can be easily purified. Most VLPs comprise at least a viral core protein that drives budding and release of particles from a host cell. One example of such a core protein is influenza M1 . Influenza VLPs can be produced by transfection of host cells with plasmids encoding the HA and NA proteins, and optionally the M1 protein. After incubation of the transfected cells for an appropriate time to allow for protein expression (such as for approximately 72 hours), VLPs can be isolated from cell culture supernatants. In another embodiment, the variants comprised in the vaccine composition are comprised in a scaffold. An especially preferred scaffold is a virosome. A virosome is a particle in which a membrane such as a phospholipid membrane encloses the virus derived proteins. The membrane may comprise viral surface antigens such as HA and/or NA. Virosomes are generally useful for protein delivery to target cells by membrane fusion.

In still another embodiment, the variants are comprised in a vector. For example, the influenza virus antigens may be expressed by said vector upon delivery to host cells. The vector may be DNA-based or RNA-based. In a preferred embodiment, the vector is a viral vector.

The concept of including adjuvants in vaccine compositions is well known in the field of vaccination. The presence of adjuvants in a vaccine composition may allow to minimize the amount of antigenic material to be administered while maintaining stimulation of an immune response sufficient for inducing immunity against the antigenic material, e.g., by inducing generation of specific antibodies. In various embodiments of the present invention, the vaccine composition may further comprise one or more adjuvants. An "adjuvant" shall mean any agent suitable for enhancing the immunogenicity of an antigen and boosting an immune response in a subject. Numerous adjuvants, including particulate adjuvants, suitable for use with both protein- and nucleic acid-based vaccines, and methods of combining adjuvants with antigens, are well known to those skilled in the art. Adjuvants suitable for use with protein immunization include, but are not limited to, alum, Freund's incomplete adjuvant (FIA), saponin, Quil A, and QS-21 . Immunostimulatory oligonucleotides (such as those including a CpG motif) can also be used as adjuvants. Adjuvants also include biological molecules, such as costimulatory molecules. Exemplary biological adjuvants include IL-2, ANTES, GM-CSF, TNF-a, IFN-γ, G-CSF, LFA-3, CD72, B7-1 , B7-2, OX-40L and 4-1 BBL. The vaccine compositions according to the present invention may be administered by any known route of administration. Exemplary routes of administration of a vaccine composition of the invention include oral, transdermal, and parenteral delivery. Suitable routes of administration may, for example, include depot, oral, rectal, transmucosal, or intestinal administration; parenteral delivery, including intramuscular, subcutaneous, intravenous, intramedullary injections, as well as intrathecal, direct intraventricular, intraperitoneal, intranasal, or intraocular injections.

The vaccine compositions according to the present invention are useful in prevention and/or treatment of influenza infections. As detailed above, the vaccine composition according to the present invention can be used in a method of generating an immune response against influenza virus in a subject.

Specifically, the vaccine composition can be used in a method of effecting cross- neutralization against at least one variant of said subtype of said influenza A and/or B virus antigen which is different from said variants comprised in said vaccine.

Preferably, the variant of a subtype of an influenza A virus antigen against which cross- neutralization in an influenza microneutralization assay is effected differs from each variant of a subtype of an influenza A virus antigen comprised in the vaccine composition by at least 20 amino acids over their entire length. Said variant of a subtype of an influenza A virus antigen against which cross-neutralization in an influenza microneutralization assay is effected may also differ from each variant of a subtype of an influenza A virus antigen comprised in the vaccine composition by at least 25, 30, 40, 50, 60, 70, 80, 90 or even 100 amino acids. However, it may not differ by more than 120 or preferably 1 10 amino acids.

Preferably, the variant of a subtype of an influenza B virus antigen against which cross- neutralization in an influenza microneutralization assay is effected differs from each variant of a subtype of an influenza B virus antigen comprised in the vaccine composition by at least 15 amino acids over their entire length. Said variant of a subtype of an influenza B virus antigen against which cross-neutralization in an influenza microneutralization assay is effected may also differ from each variant of a subtype of an influenza A virus antigen comprised in the vaccine composition by at least 16, 18, 20, 25, 30, 35, 40 or even 50 amino acids. However, it may not differ by more than 70 or preferably 60 amino acids.

The "immune response" is preferably a "protective" immune response. A "protective" immune response refers to the ability of a vaccine to elicit an immune response, either humoral or cell mediated or both, which serves to protect the subject from influenza. The protection provided need not be absolute, i.e., influenza need not be totally prevented or influenza viruses be totally eradicated, if there is a statistically significant improvement compared with a control population of subjects. Protection may be limited to mitigating the severity or rapidity of onset of symptoms of influenza. The immune response is preferably sufficient to treat and/or prevent influenza. When used herein, the terms "treating" or "preventing" influenza denote at least an inhibition of replication of the causative influenza virus, inhibition of influenza transmission, prevention of an influenza virus from establishing itself in a subject, and/or amelioration or alleviation of the symptoms of the influenza infection.

Also, the present invention provides the use of at least three variants of a subtype of an influenza A virus antigen, wherein

(i) a first and second variant of a subtype of an influenza A virus antigen differ from each other by at least 10 amino acids over their entire length;

(ii) a second and third variant of a subtype of an influenza A virus antigen differ from each other by at least 10 amino acids over their entire length; and

(iii) a first and third variant of a subtype of an influenza A virus antigen differ from each other by more than 35 amino acids over their entire length

for the manufacture of a vaccine composition which effects cross-neutralization against at least one variant of said subtype of said influenza A virus antigen which is different from said variants comprised in said vaccine.

Moreover, the present invention provides the use of at least three variants of a subtype of an influenza B virus antigen, wherein

(i) a first and second variant of a subtype of an influenza B virus antigen differ from each other by at least 10 amino acids over their entire length;

(ii) a second and third variant of a subtype of an influenza B virus antigen differ from each other by at least 10 amino acids over their entire length; and

(iii) a first and third variant of a subtype of an influenza B virus antigen differ from each other by more than 20 amino acids over their entire length for the manufacture of a vaccine composition which effects cross-neutralization against at least one variant of said subtype of said influenza B virus antigen which is different from said variants comprised in said vaccine. The present invention provides a method for producing a vaccine composition against influenza A virus, comprising

(a) selecting at least three variants of a subtype of a variant of an influenza A virus antigen, wherein

(i) a first and second variant of a subtype of an influenza A virus antigen differ from each other by at least 10 amino acids over their entire length;

(ii) a second and third variant of a subtype of an influenza A virus antigen differ from each other by at least 10 amino acids over their entire length; and

(iii) a first and third variant of a subtype of an influenza A virus antigen differ from each other by more than 35 amino acids over their entire length

(b) combining said selected variants in one vaccine composition; and

(c) optionally determining whether said vaccine composition effects cross-neutralization against at least one variant of said subtype of said influenza A virus antigen which is different from said variants comprised in said vaccine.

The present invention provides a method for producing a vaccine composition against influenza B virus, comprising

(a) selecting at least three variants of a subtype of a variant of an influenza B virus antigen, wherein

(i) a first and second variant of a subtype of an influenza B virus antigen differ from each other by at least 10 amino acids over their entire length;

(ii) a second and third variant of a subtype of an influenza B virus antigen differ from each other by at least 10 amino acids over their entire length; and

(iii) a first and third variant of a subtype of an influenza B virus antigen differ from each other by more than 20 amino acids over their entire length

(b) combining said selected variants in one vaccine composition; and

(c) optionally determining whether said vaccine composition effects cross-neutralization against at least one variant of said subtype of said influenza B virus antigen which is different from said variants comprised in said vaccine.

The embodiments described above under the headings "First, second and third variant of a subtype of an influenza A virus antigen", "First, third and fourth variant of a subtype of an influenza A virus antigen", "First, fourth and fifth variant of a subtype of an influenza A virus antigen" and "Sixth, seventh, eighth, ninth or tenth variant of a subtype of an influenza A virus antigen", "First, second and third variant of a subtype of an influenza B virus antigen", "First, third and fourth variant of a subtype of an influenza B virus antigen", "First, fourth and fifth variant of a subtype of an influenza B virus antigen" and "Sixth, seventh, eighth, ninth or tenth variant of a subtype of an influenza B virus antigen" also apply to the herein described uses for the manufacture of a vaccine composition which effects cross-neutralization and methods for producing a vaccine composition, respectively.

EXAMPLES

A. Generation of Influenza VLPs

Generation of the constructs

All genes were individually cloned in a vector in order to have fully functional expression cassettes comprising a promoter, the open reading frame and a terminator compatible with the insect cell-baculovirus expression vector system (IC/BEVS). The open reading frames were synthesized by Genescript with additional flaking regions and a Pmel restriction site which facilitated their cloning using a standard Gibson protocol. Those expression cassettes were afterwards ligated in another cloning vector in various combinations in order to have the expression of VLPs carrying up to five hemagglutinis or neuraminidases to be expressed using the IC/BEVS. An overview of the constructs for expression of VLPs is shown in Figure 7.

The final gene combinations were then shuffled into a vector harbouring the recognition sites for the Tn7 bacterial transposase for generation of recombinant baculoviruses for expression in insect cells. In this context, the whole shuffled cassette comprising one or more genes was inserted in the viral genome via recombination using the bacmid Tn7 site, as previously described (Luckow et al. (1993), J Virol. 67(8),:4566-4579).

The following H1 variants were used for the generation of VLPs:

A/Puerto Rico/8/1934 (SEQ ID No: 1 ),

A/USSR/92/1977 (SEQ ID No: 2),

A/Texas/36/1991 (SEQ ID No: 3),

A/New Caledonia/20/1999 (SEQ ID No: 4),

A/Brisbane/59/2007 (SEQ ID No: 5).

Broadening was tested against A/California/04/2009 (SEQ ID No: 6), A/California/07/2009 (SEQ ID No: 7), A/Mexico/INDRE4487/2009 (SEQ ID No: 8)

The following H3 variants were used for the generation of VLPs:

A/Hong Kong/1/1968 (SEQ ID No: 9),

A/England/321/1977 (SEQ ID No: 10),

A/Sichuan/2/1987 (SEQ ID No: 1 1 ),

A/Nanchang/933/1995 (SEQ ID No: 12) or A/Johannesburg/33/94 (SEQ ID No: 13), A/Wyoming/3/2003 (SEQ ID NO: 14) or A/Fujian/41 1/2002 (SEQ ID No: 15).

Broadening was tested against A/Switzerland/9715293/2013 (SEQ ID No: 16), A/Hong Kong/4801/2014 (SEQ ID No: 17, 18, 19 or 20).

The following HB variants were used for the generation of VLPs: B/Hong Kong/8/73 (SEQ ID No: 21 ),

B/Beijing/1/1987 (SEQ ID No: 22) or B/Victori a/02/1987 (SEQ ID No: 23),

B/Yamagata/16/1988 (SEQ ID No: 24),

B/Jiangsu/10/2003 (SEQ ID No: 25),

B/Malaysia/2506/2004 (SEQ ID No: 26).

Broadening was tested against B/Brisbane/60/2008 (SEQ ID No: 27 or 28), B/Phuket/3073/2013 (SEQ ID No: 29 or 30).

The following H4 variants can be used for the generation of VLPs:

A/Mallard/Alberta/47/1991 (SEQ ID No: 34),

A/cinnamon teal/Bolivia/4537/2001 (SEQ ID No: 35),

A/quail/N J/25254-22/1995 (SEQ ID No: 36),

A/herring gull/Astrakhan/267/1982 (SEQ ID No: 37),

A/Australian shelduck/Western Australia/1762/1979 (SEQ ID No: 38).

The following NA N1 variants can be used for the generation of VLPs:

A/Puerto Rico/8/1934 (SEQ ID No: 31 ),

A/Texas/36/1991 (SEQ ID No: 32),

A/Brisbane/59/2007 (SEQ ID No: 33).

Cell culture

Spodoptera frugiperda derived Sf9 were maintained in a protein-free liquid culture medium (Insect-XPRESS™) and routinely re-inoculated every 3-4 days at 0.6 to 1.0 x 10 ⁶ cells ml-1.

Transfection of insect cells

The recombinant bacmids were transfected into Spodoptera frugiperda (Sf9) host cells using the reagent ViaFect™ (Promega), according to the manufacturer protocol. The viruses were harvested three days post infection and amplified to generate the master seed virus.

B. Production process of Influenza VLPs

Cell line and culture media

Parental or stable insect High Five cells were routinely sub-cultured to 0.3x10 ⁶ cells/mL every 3-4 days when cell density reached 2-3x10 ⁶ cells/mL in 125, 250 or 500 mL shake flasks (10 % working volume) in an Innova 44R incubator at 27 °C and 100 rpm (orbital motion diameter of 2.54 cm) using Insect XPRESS™ medium. Baculovirus amplification

Recombinant baculoviruses containing Influenza matrix M1 gene alone or in combination with one or multiple hemagglutinin (HA) genes were generated as described herein (see A). Amplification of baculovirus stocks was performed as described in Vieira et al., J Biotechnol 2005;120:72-82. Briefly, Spodoptera frugiperda Sf-9 cells infected at 1 *10 ⁶ cells/mL at a multiplicity of infection (MOI) of 0.1 infectious particles per cell (ip/cell). When cell viability reached 80-85 %, cultures were harvested and centrifuged at 200 g for 10 min at 4 °C. The pellet was discarded and the supernatant was centrifuged at 2000 g for 20 min at 4 °C. The resulting supernatant was stored at 4 °C until further use.

Production of influenza VLPs

Influenza VLPs were produced in shake flaks (125, 250 or 500 mL with 10 % working volume), in glass stirred tank bioreactors or in single-use wave induced bioreactors. Briefly, shake flask cultures were infected at a cell concentration at infection (CCI) of 1 , 2, 3 or 4*10 ⁶ cells/mL using a MOI of 0.1 , 1 or 10 ip/cell. Specifically for infections performed at CCI of 4*10 ⁶ cells/mL, when cell concentration reached 2*10 ⁶ cells/mL, the culture medium was supplemented with a mixture containing insect medium supplement 10x, 5 mM glutamine, 10 mM asparagine and 20 mM glucose at a ratio of 10 % (v/v) regarding the final culture volume.

Bioreactor cultures were performed in computer-controlled BIOSTAT ^® B-DCU 2L vessels equipped with two Rushton impellers, a sparger for gases supply, a water recirculation jacket for temperature control, and multiple ports for temperature, pH, p0 ₂ (partial pressure of oxygen) probes as well as for additions (e.g. culture medium) and sampling/harvesting of cell culture. The p0 ₂ was set to 30 % of air saturation and maintained by varying the agitation rate from 70 to 250 rpm and the percentage of 0 ₂ in the gas mixture from 0 to 100 %. The gas flow rate was set to 0.01 wm and temperature was kept at 27 °C. The working volume was 2 L. All cultures were inoculated at a cell density around 0.5* 10 ⁶ cell/ml and allowed to grow up to 2 or 4x10 ⁶ cell/ml (CCI) before infection with baculoviruses at a MOI of 1 ip/cell. Specifically for cell cultures infected at CCI of 4x10 ⁶ cells/mL, a supplementation scheme similar to that described above was adopted.

Bioreactor cultures were also performed in computer-controlled wave induced bioreactors, Wave™ 20/50 EH, equipped with multiple ports for temperature and p0 ₂ probes as well as for additions (e.g. culture medium) and sampling/harvesting of cell culture. The p0 ₂ was set to 30 % of air saturation and controlled by varying the agitation rate and sequentially the percentage of N ₂ and 0 ₂ in the gas mixture between 0-100 %. The gas flow rate was set to 0.03 vvm. The temperature was kept at 27 °C by using an in-situ heating plate supporting the wave bag. The working volume was 10 L. All cultures were inoculated at a cell density around 0.5*10 ⁶ cell/ml and allowed to grow up to 2*10 ⁶ cell/ml (CCI) before infection with baculoviruses at a MOI of 0.1 -1 ip/cell.

Cell cultures were harvested when cells viability dropped to values around 50%, commonly between 48h post infection (hpi) and 96 hpi, and processed immediately according to the purification schemes described in C.

C. Purification process of Influenza VLPs

The downstream processing (DSP) scheme used for purification of VLPs consisted in a five- stage process that included a clarification step for cells removal, an UF/DF step for product concentration (TFF-membrane cassette), a chromatographic step for product purification (AEX or SEC), a second UF/DF step for final product concentration, and a final sterile filtration step (0.2 pm filter) before final formulation.

Briefly, two clarification schemes were used for cells removal: Scheme 1 - the supernatant was passed through two depth filters (5 pm and 0.65 pm), or Scheme 2 - culture broth was centrifuged at 200 g for 10 min at 4 °C, supernatant was collected, mixed with Benzonase (50 U/mL) for digestion of nucleic acids and then passed through two depth filters of 0.45 pm and 0.2 pm.

Two UF/DF schemes were used for product concentration: Scheme 1 - clarified bulk was passed through a cassette with 500 kDa of pore size keeping TMP within 0.8-1 bar and permeate flux between 25-35 LMH, or Scheme 2 - clarified bulk was passed through a cassette 300 kDa of composite regenerated cellulose keeping TMP within 0.8-1 bar, 24 LMH permeate flux and retention flux of 48 LMH membrane area.

Two modes of chromatography were used for product purification: Scheme 1 - retentate of UF/DF step was passed through an AEX chromatographic column using Buffer A (50 mM HEPES pH 7.5), Buffer B (Buffer A + 1 M NaCI), starting conductivity of 15 mS/cm, 25 CV gradient length and a flow rate of 2.5 CVs/mL, or Scheme 2 - retentate of UF/DF step was passed through a SEC column using a buffer containing 50 mM HEPES, pH 7.5 and 300 mM of NaCI, a volume injection of 5% CV and a flow rate 4 mL/min.

Two UF/DF schemes were used for final product concentration: Scheme 1 - the purified product coming from AEX or SEC was concentrated using hollow fibers 750 kDa PES, 50 cm ² , operated at TMP lower than 0.8 bar (measured with the two pressure sensors placed on the column valve of an AKTA avant) and 30 mL/min recirculation flow rate, or Scheme 2 - the purified product coming from AEX or SEC was concentrated using 300kDa cassette regenerated cellulose of 50 cm ² operated at TMP of 1.2 bar, 40 ml/min and 5DV - final formulation. The final sterile filtration step was performed using two different filters, the Acrodisc from Pall Lifesciences and the Whatman cellulose regenerated membrane filter. Purified VLPs were formulated in a buffer consisting of 50 mM HEPES, 300 mM NaCI, pH 7.4 and trehalose 15% (w/v), and stored at -80 °C until further. D. Analytical for product characterization

Cell concentration and viability

Cell concentration, viability, size and volume were monitored daily using (i) the trypan blue exclusion dye method with a Fuchs-Rosenthal haemocytometer, (ii) the Casy ^® 1 Cell counter plus Analyzer System Model TTC, or (iii) the Cedex High Resolution Cell Analyzer.

Metabolite analysis

Glucose, glutamine, lactate and glutamate concentrations were determined with automated enzymatic assays using the YSI 7100 Multiparameter Bioanalytical System. The concentration of other metabolites was estimated by ¹ H-NMR spectroscopy as described in Carinhas et al., Biotechnol Bioeng 2013; 1 10:3244-3257. Briefly, spectra were recorded in a 500 MHz Avance spectrometer equipped with a 5 mm QXI inversed probe, using a NOESY- based pulse sequence with water pre-saturation. DSS-d6 was used as internal standard for metabolite quantification in all samples. In order to obtain a similar pH between samples, they were mixed with phosphate buffer (pH 7.4) prepared in D ₂ 0 at a 2:1 ratio. Each spectrum was phased, baseline corrected and integrated using the Chenomx NMR Suite 8.0 software.

Transmission electron microscopy (TEM)

Negative staining TEM was used to assess the conformation and size of purified Influenza VLPs. Briefly, 10 μΙ of purified VLP sample was fixed for 1 min in a copper grid coated with Formvar-carbon. Grids were washed with H ₂ 0 and then stained with 1 % uranyl acetate for 2 min and left to air dry. Samples were then observed in a Hitachi H-7650 Transmission Electron Microscope.

Western blot

Western blot was used to confirm the presence of Influenza proteins in monovalent and multivalent VLPs. Proteins were first denatured by heating the sample to 70 °C for 10 min, then separated under reducing conditions in a 4-12 % SDS gel and finally transferred to a nitrocellulose membrane using iBIot ^® Transfer Stack. Several antibodies received from NIBSC and Influenza Reagent Resource (IRR) (established by the Centers for Disease Control and Prevention) were used for HA identification. The IRR antibody FR-494 was used to identify HA in monovalent and multivalent H1 VLPs. The NIBSC influenza anti- A/Johannesburg/33/1994 and anti-A/Nanchang/933/1995 HA sera were used to detect HA in monovalent and multivalent H3 VLPs. The NIBSC influenza anti-B/Hong Kong/8/73 and anti- B/Victoria/2/87 HA sera were used to detect HA in monovalent and multivalent B VLPs. For M1 protein identification, a comercially available antibody was used. All primary antibodies were used at dilutions between 1 :1000 and 1 :2000. As secondary antibodies, anti-sheep, anti-mouse or anti-goat antibodies conjugated with either HRP-labeling or AP were used at dilutions between 1 :2000 and 1 :5000. Protein band detection was performed using the enhanced chemiluminescence detection system or the 1-step™ NBT/BCIP blotting detection reagents. Single radial immunodiffusion (SRID)

The SRID protocol was used to estimate the concentration of immunologically active HA in purified Influenza VLPs. This technique is based on the ability of antigen forming a precipitate ring when in contact with equal concentration of antiserum. Briefly, antiserum was added to an agarose solution of 1 % (w/v) before the gel was made. The quantity of antiserum to mix with agarose depended on the number of plates to perform and reagent specifications. After gels have set, wells were cut into the agar gel and plates were left at 4°C (humidified) until further use. Antigen (standard or purified VLP) was treated with Zwittergent 3-14 detergent 10 % (w/v) and the mixture was left at RT for 30 min. Twenty (20) μΙ of each antigen solution was added into the wells cut into the agar gel and then the plates were incubated for a minimum of 18 h in a moist box at 20-25 °C. During this time, the antigen was able to diffuse from wells; as the antigen diffused out of the well, its concentration still remained above that of antiserum thus forming relatively soluble antigen- antibody (Ag-As) adducts. As the antigen diffused further and further from the well, its concentration decreased up to the point where it matched that of the antiserum in the gel. At that point, an Ag-As precipitate ring was formed. The procedure to reveal this Ag-As precipitate ring is depicted in Wood et al., Journal of Biological Standardization, Volume 5, Issue 3: 237 -247. Briefly, plates were washed for 30 min in PBS, a moist filter paper was added on the agarose surface followed by four layers of absorbent material, and the gel was then pressed for 30 min. The weight was removed and, with the filter paper still attached, plates were allowed to dry in warm air. When plates and filter paper were dry, the filter paper was removed and plates stained with Coomassie Brilliant Blue 0.3 % (w/v) in a mixture of acetic acid (12 %) and methanol (29 %) for 5-10 min. Finally, the plates were destained using a mixture of acetic acid (12 %) and methanol (29 %) for 15-30 min, allowed to dry in warm air, and the diameter of the Ag-As rings measured in different angles. By running a standard of known concentrations along samples for which antigen concentration is unknown on the same gel, and comparing their precipitate rings diameter, one could estimate samples antigen concentration.

Bicinchoninic acid (BCA) protein assay

The principle of the BCA assay relies on the formation of a copper-protein complex under alkaline conditions, followed by reduction of Cu ²⁺ to Cu ⁺ . The extent of copper ions reduction is proportional to the amount of protein present. Since BCA forms a purple-blue complex with Cu ⁺ in alkaline environments, it is possible to estimate the concentration of total protein in a sample by comparing the color development of that sample with a standard of known concentration. Total protein content in purified VLP samples was determined using the BCA protein assay kit (96-well plate protocol) from Pierce Biotechnology following manufacturer's protocol. AccuBlue assay

The AccuBlue™ assay allows precise quantitation of purified dsDNA across a wide range of concentrations by using fluorescent DNA binding dyes that are highly sensitive and selective for dsDNA. Within the variety of AccuBlue dsDNA quantification kits commercially available, the AccuBlue™ Broad Range dsDNA Assay was used according to manufacturer's instructions for estimating the content of residual DNA in purified VLPs samples.

Endosafe ^® -PTS™ quantitative LAL method

Endotoxin content was quantified using Endosafe ^@ -PTS ^™ following manufacturer's instructions. The Endosafe ^® -PTS ^™ is a rapid, point-of-use test system that uses LAL reagents in an FDA-licensed test cartridge with a handled spectrophotometer. The PTS™ uses the LAL kinetic chromogenic methodology to measure color intensity (color development take up to 15 min) that is directly correlated with the concentration of endotoxins in a sample. By comparing sample's color with that of a control standard endotoxin, it is possible to estimate the concentration of endotoxins. Using this analytical tool the concentration of endotoxins in purified VLPs samples was estimated.

Sterility tests, pH and Mycoplasma detection

The fluid thioglycollate medium (a general-purpose medium for the cultivation of anaerobes, microaerophiles and aerobes) and the soybean casein digest broth (a general-purpose medium for the cultivation of bacteria and fungi) are formulations adopted by the United States Pharmacopeia and the European Pharmacopeia as sterility test media. Therefore, they were herein used to confirm the sterility of purified VLPs samples. Briefly, 10-300 μΙ of purified samples was added to 5 ml of both media and the mixture allowed to incubate at 27 or 37 °C for a maximum of 3 or 5 days. After incubation, growth of contaminants is evidenced by the presence of turbidity in the tubes. The pH of in-process and purified VLP samples was assessed using the Crison Micro pH 2002 system. Master baculovirus seed stocks were screened for the presence of mycoplasmas using qPCR.

Nanoparticle tracking analysis

NanoSight NS500 was used for nanoparticle tracking analysis throughout the DSP of Influenza VLPs, and thus estimate total particle concentration in purified VLPs samples. It utilizes the properties of both light scattering and Brownian motion to obtain the size distribution and concentration measurement of particles in solution. By laser beaming particles in solution when these pass through a sample chamber, particles scatter light in such a manner that they can be visualized via a magnification microscope onto which a camera is mounted. Then, a software analysis the video file of the particles moving under Brownian motion recorded by this camera and using the Stokes-Einstein equation calculates their hydrodynamic diameters. Briefly, the samples were diluted in D-PBS to a concentration of 10 ^s — 10 ⁹ particles ml_ ^"1 , the instrument's linear range. All measurements were performed at room temperature (22 °C). Sample videos were analyzed with the Nanoparticle Tracking Analysis (NTA) 2.3 Analytical software, release version build 0025. Capture settings (shutter and gain) were adjusted manually. For each sample 60-seconds videos were acquired and particles between 70 and 150 nm were considered.

Baculovirus titration

Baculovirus titers were assessed using the MTT assay, a colorimetric assay that measures the reduction of yellow 3-(4,5-dimethythiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) to an insoluble, coloured (dark purple) formazan product by NAD(P)H-dependent cellular oxidoreductase enzymes. Cells are then solubilised with dimethyl sulfoxide and the released, solubilized formazan crystals are measured spectrophotometrically. Since only metabolically active cells can reduce MTT, the development of purple color is directly correlated with the number of viable cells. Because the baculovirus is a lytic virus, infection will reduce cell growth. Such a reduction is dose-dependent and can be estimated by measuring the viable cell concentration using MTT and can be correlated to the viral titer (i.e. concentration of infectious baculovirus) using a set of mathematical regressions(Roldao et al., J Virol Methods 2009; 159:69-80). Briefly, 100 μΙ of 5x10 ⁵ cell/ml of Sf-9 cells were seeded onto a 96-well tissue culture plate and allowed to settle for 1 h at 27 °C. Supernatant was removed and 100 μΙ of viral stocks diluted 10 ^~1 to 10 ^~10 times was added per well and plates incubated 6 days at 27 °C. MTT was added (10 % (v/v) of total volume per well at 5 mg/ml) and plates incubated at 27 °C for additional 4 h. The supernatant was removed and formazan crystals solubilised by adding 150 μΙ per well of DMSO. Plates were agitated for 10-20 min in a wellmix shaker WM-506 and absorbance (570/690 nm wavelength) measured using Infinite ^® 200 PRO NanoQuant microplate reader. Collected data was analyzed using Prism 5 for Windows to determine the tissue culture lethal dose 50 (TCLD ₅₀ ). The conversion of TCLD ₅₀ to viral titers (pfu/ml) was carried out using the mathematical regressions reported elsewhere ⁴ .

Enzyme-linked immunosorbent assay (ELISA)

ELISA is a plate-based assay technique designed to detect the presence of an antigen and estimate its concentration. The ELISA herein used for detection of HA protein was an indirect ELISA and consisted in: (1 ) immobilization of the antigen to a solid surface, (2) incubation with a primary antibody followed by a secondary antibody that is linked to horseradish peroxidase, (3) incubation with the TMB substrate (3,3',5,5'- Tetramethylbenzidine) to produce a measureable product, and (4) signal-detection using a spectrophotometer. Briefly, cell culture samples were centrifuged at 200 g, 4 °C for 10 min and supernatants collected for ELISA analysis. A mixture of 1 :10 or 1 :100 of culture supernatant in coating buffer (0.1 M Na ₂ HP0 ₄ ) was added to a 96 well Nunc-Maxisorp plate (100 μΙ/well) and allowed to incubate overnight at 4 °C. Plate was washed 3x with wash buffer (PBS with 0.05 % Tween-20) and blocked with 200 μΙ/well blocking buffer (3 % BSA in PBS) for 1 h at room temperature. Then, 100 μΙ/well of primary antibody in dilution buffer (3 % BSA in PBS with 0.05 % Tween-20) was added and plate incubated for 1 h at RT. The plate was washed 3x with wash buffer and 100 μΙ/well of secondary HRP-coupled antibody in dilution buffer was added. After 1 h at RT, plate was washed 3x with wash buffer and developer TMB Substrate (100 μΙ/well) was added. Reaction was stopped with 100 μΙ 1 M HCI as soon as positive wells were blue and the plate was subsequently read at 450nm using a Infinite ^® 200 PRO NanoQuant microplate reader. Hemagglutination assay

Hemagglutination assay is based on the process of hemagglutination, in which sialic acid receptors on the surface of red blood cells (RBCs) bind to HA localized on the surface of influenza viruses, creating a network (lattice structure) of interconnected RBC and viral particles (Hirst (1942),. J Exp Med 75: 49-64). The formation of this lattice structure is correlated with the concentration of viral particles in a specific sample. For example, if the virus concentration is too low, there is not enough HA to bind to RBC and thus RBC settle to the bottom of the well. The RBC used in this assay are typically from chickens, turkeys, horses, guinea pigs and humans but are highly dependent on the selectivity of the targeted virus and the associated surface receptors on the RBC. The hemagglutination assay herein used is a plate-based assay in which the concentration of HA in bulk and purified VLP samples is determined by comparing the hemagglutination profile of these samples with that of a standard of known HA concentration. Briefly, samples were 2-fold serially diluted in PBS and incubated for 30 min at 4 °C with 25 μΙ of 1 % chicken RBC. Hemagglutination of RBC was identified visually by the formation of a network (lattice structure) of interconnected RBC and HA (positive results). As standard, an influenza vaccine with a known HA concentration was added to each assay experiment. The HA titer of each sample was determined by calculating the maximum dilution with a positive outcome and comparing it to the one obtained for the standard. Protein deglvcosylation

The alteration of proteins by post-translational modifications (e.g. glycosylation) generates a wide range of molecules that can have an identical underlying protein core but completely different biological roles. Understanding the glycosylation pattern of proteins is therefore critical to clarify the nature of the numerous variants observed and its potential impact on cellular processes. Target-specific enzymes (e.g. trifluoromethanesulphonic acid) (Edge (2003), Biochem J. 376(Pt 2): 339-350) are commonly used to selectively remove carbohydrates from glycoproteins, while leaving the protein backbone intact, and thus assist the elucidation of protein structure and function. Besides protein structural/functional studies, protein deglycosylation can be very useful to improve between-laboratory calibration of influenza antigen standards (Harvey et al. (2012), Biologicals 40(1 ): 96-99.). The protein deglycosylation protocol was kindly provided by NIBSC. Briefly, supernatant of culture samples or purified VLPs samples were mixed with 2 μΙ of 10x denaturing buffer (provided with enzyme) up to a total volume of 14 μΙ (add water if needed) and incubated at 95-100 °C for 10 min. 7 μΙ of denatured samples were mixed with 1 μΙ of 10x G7 buffer (provided with enzyme), 1 μΙ of 10 % NP40 (provided with enzyme) and 1 μΙ of 1/20 dilution of PNGase F (or water for a non-deglycosylated control), and incubated at 37 °C overnight. Deglycosylated protein samples were then assessed by SDS-PAGE followed by Western blot analysis as described above.

Mass spectometry

Mass spectrometry was used for identification of HA and M1 proteins in bulk and purified VLP samples. Briefly, samples were run in SDS-PAGE gels, stained with coomassie blue and gel bands corresponding to HA or M1 proteins (according to Western blot results) were excised, digested (trypsin), desalted and concentrated using C18 microcolumns. Eluates were directly spotted on a MALDI plate (matrix used was a-Cyano-4-hydroxycinnamic acid) and analysed on a 4800 Plus MALDI-TOF/TOF analyser. Raw data were generated by the 4000 Series Explorer Software v3.0 RC1 . Database search for protein identification was performed using the algorithm MOWSE (version 2.2). Swissprot database and a custom database extracted from SwissProt containing extra HA entries (to include all HA and M1 protein sequences herein used) were used.

Isotope dilution mass spectrometry

An Isotope Dilution Mass Spectrometry (IDMS) method was herein developed for quantitation of HA in the purified monovalent and multivalent VLPs. Briefly, samples were denatured and digested with trypsin. Isotopically labelled peptides were added in the digest as internal standards (IS). The level of HA peptides was determined following peptide standard curves, where the peptide fragment ion intensity was used and expressed as the ratio of the native peptide to the heavy peptide (IS). The protein amount was calculated based on the molarity of peptides multiplying by HA molecular mass.

Immunofluorescence microscopy

An immunofluorescence protocol was used to visually inspect the presence of HA in insect cell membranes (Correia et al., Stem Cells Transl Med 2016;5:658-69). Briefly, culture samples containing 2*10 ⁶ cells were centrifuged at 300 g for 5 min, the cell pellets were collected, washed with PBS twice and then incubated with 50 μΙ of anti-HA antibody solution (dilution of 1 :20 in PBS) at 4 °C in the dark for 1 h. Afterwards, samples were centrifuged, the cell pellets were washed twice with PBS and then incubated with 50 μΙ of secondary antibody Alexa Fluor ^® 488 (dilution 1 :200) for 30 min at 4 °C in the dark. After two washing steps with PBS, samples were re-suspended in PBS and fluorescence microscopy analysis was performed to detect GFP.

Real-time quantitative PCR

Real-time quantitative PCR was used to detect and quantify a targeted DNA molecule (the ie-1 gene) from baculoviruses as described in (Vieira et al. (2005), J Biotechnol, 120(1 ):72- 82.). Using this analytical technique we were able to estimate the concentration of total baculovirus particles in in-process and purified VLP samples. E. Mice immunogenicity studies

Materials and Methods

Preliminary mouse studies focused on the generation of high titre reference sera from monovalent H1 , H3, NA and B strains VLPs for use in the project and for harmonisation purposes. In parallel, pilot mouse studies have been conducted on monovalent H1 , H3, NA and B strains VLPs, as well as with a mixture of all variants per (sub)type, to asses assay conditions and verify the vaccine dose and immunisation regimen.

Moreover, mouse vaccination studies with seasonal influenza vaccine and with HA/NA VLPs mix to exclude antigen competition has been conducted.

The VLPs generated as described in A and B have been evaluated in outbred (Swiss) mice. The total HA dose per vaccination was kept at 1 .5 μg (1/10th of a human dose per (sub) type), and 0.5 μg for the NA, both irrespective of the VLPs valence. As the mice are influenza naive, in contrast to the human target population, we vaccinated three times at a four weekly intervals to allow the humoral immune response to mature. Generation of reference mouse sera

In order to obtain sufficient quantities of mouse high titer sera from each of the 20 selected monovalent HAs (H1 , H3, B) and 3 NAs, groups of four outbred Swiss mice, six-weeks old female, have been immunised subcutaneously three times with 1.5 μg HA or 0.5 μg NA adjuvanted with Montanide ISA 51 VG, Seppic, France, at four week intervals. All vaccine formulations have been performed in a laminar flow cabinet mixing antigen solution and ISA 51 VG in 1 :1 ratio. Homogeneous water in oil emulsion was reached passing antigen solution and adjuvant 20 times back and forth through a 22-gauge couple piece. Emulsion formed and viscosity increased during process. Formulations were inspected under phase contrast microscopy, at 1000 x magnification. Three bleeding points, at day 0 to confirm flu sera negativity, at day 42 for intermediate evaluation and at day 70 for the final bleeding have been performed. On average, 1.2 mL serum/pool has been obtained for each strain. Pooled sera from mice immunised with the five selected HA variants from each subtype (H1 , H3, B) and with the three selected NA variants, have been analysed for specific HA / NA IgG quantity using ADAMSEL FPL, a free-software for non-commercial users. This application converts the Optical Density (OD) readings obtained from ELISA plate readers into concentrations by four-parameter fitting. This software is designed to provide an auditable system that minimises data handling, thereby reducing the chances of error. With this system, a yield of 1 OD over background has been considered as 1 AU/mL.

Pilot mouse study

Exploratory experiments have been conducted with all five HA variants per (sub)type (H1 , H3, B) and with the three selected NA variants to define the vaccine dose/regimen and to verify, compare and establish kinetics as well as assay conditions. Groups of five outbred Swiss mice, six-week old female, have been subcutaneously immunised with monovalent HA VLPs (without adjuvant) as well as with a mixture of all five variants per (sub) type (H1 , H3, B). The total HA dose per vaccination was kept at 1.5 μg. Groups of five Swiss mice, six- week old female, have been vaccinated with monovalent NA, as well as with a mixture of three N1 VLPs at a 0.5 pg NA dose.

Three bleeding points, at day 0 to confirm flu sera negativity, at day 42 for intermediate evaluation and at day 70 for the final bleeding have been performed. The sera generated were individually collected (300 μΙ sera/mouse) and used for ELISA testing.

Mouse vaccination study: seasonal influenza vaccine

The aim of this study was:

1. To establish that immunisation with polyvalent VLPs expressing different HA antigens from the same subtype will induce broad protection within that subtype.

2. To determine the minimum number of components required.

3. To establish that multivalent VLPs of H1 , H3 and B can be administered simultaneously without loss of immunogenicity and neutralising capacity.

Groups of eight outbred Swiss mice, six-week old female, have been immunised with 1.5 μg HA of monovalent, trivalent and pentavalent VLPs for each sub type (H1 , H3 and B) as well as with a mixture of H1 , H3 and B subtypes (4.5 pg total HA/dose). The selected monovalent strains used for the mixture are the following: A Texas/36/1991 (H1 variant 3), A/Sichuan/2/1987 (H3 variant 3) and B/Yamagata/16/1988 (B variant 3).

For HA/NA competition studies, eight mice have been immunised with 0.5 pg of the trivalent NA VLPs (N1 -2-3) and eight mice with the mix of trivalent NA / H3 VLPs. Three bleeding points, at day 0 to confirm flu sera negativity, at day 42 for intermediate evaluation and at day 70 for the final bleeding have been performed. The sera generated were individually collected (300 μΙ sera/mouse) and used for ELISA testing. Mouse vaccination study: hexavalent group 2 HA vaccine

Hexavalent group 2HAs VLPs (H3, H4, H7, H10, H14, H15) can be similarly assessed, separately and in combination, to investigate the potential to induce heterosubtypic neutralising antibodies.

Groups of eight outbred Swiss mice, six-week old female, can been immunised with 1.5 μg HA of each of the six monovalent of group 2HA and with the mix of them, as well as with 1.5 μg of the hexavalent VLP.

Three bleeding points, at day 0 to confirm flu sera negativity, at day 42 for intermediate evaluation and at day 70 for the final bleeding can be performed. The sera generated can be individually collected (300 μΙ sera/mouse) and used for ELISA testing.

Production of egg-derived whole antigens

To reduce potential cross reactivity in ELISA that can arise if coating and immunising antigens are the same, antigens from a different expression system were produced to coat the ELISA plates. Egg-derived completely purified viruses have been produced for the five variants of each sub type H1 , H3 and B to be used as coating antigens for all planned ELISA assay.

Whole inactivated viruses for the five different variants of H1 N1 , H3N2 and B sub-types have been produced so far and HA concentration was determined by Single Radial Diffusion (SRD).

An average of 100 eggs was used for each variant of each subtype (H1 N1 , H3N2 and B). Seeds viruses were reconstitute at 10 ^"4 dilution. 0.2mL/egg were inoculated in fertilised chicken eggs (pre-incubated for 10-1 1 days), the allantoic fluid was harvested and cold precipitation was achieved by storing the harvest at 2-8 °C for 48 hrs.

Harvest was then clarified by refrigerated centrifugation to remove cellular debris, filtered through a 0.2 μιτι filter and desalted and concentrated by dia-filtration. The whole viruses were then purified by zonal centrifugation on a density gradient, concentrated and finally inactivated with beta-propiolactone. About 6 mg of each of the five virus variants for each subtype (H1 N1 , H3N2 and B) have been produced under Good Manufacturing Practice (GMP) conditions. SRD has been used to quantify the HA amount using the relative reagents provided by NIBSC.

Micro ELISA SOP for half area plate

Half area microplate (Greiner Bio On) allows up to a 50% sample and reagent reduction

Reagents: - Coating Buffer: 50 mM Na2CC ^< 3 pH 9.6

- PBS

- PBS/0.05% Tween 20: (500 μΙ tween 20 in 1 L PBS)

- Blocking Buffer: PBS plus 10% milk powder

- Dilution Buffer : PBS/0.05%Tween plus 3% milk powder

Day 1 :

Coat the plates with 50ng (50μΙ of a Stock solution 1 μg/mL) of the proper antigens diluted in Coating Buffer. Leave the plates at 4°C O/N

Day 2:

Remove the coating antigens by inverting the plates

- Add 100 μΙ Blocking Buffer

Incubate the plates at 37°C for 1 hr

- In the meantime, dilute animal sera in dilution buffer (start with 1 : 1000 dilution). Add 100 μΙ of diluted sera at the first column (A1 -H 1 ) and then make 1 :2 serial dilution by transferring 50 μΙ from the first column to the second one (1 to 10), and so on.

After incubation with the Blocking Buffer, wash ones (with the machine, 200 μΙ volume) with PBS.

- Add 50 μΙ Dilution Buffer in the all plate except in the first column (A1 -H1 )

- Add 100 μΙ of diluted (1 :1000) sera from A1 to H1

Make 1 :2 serial dilution from 1 to10

- Incubate the plates 1 hr at 37°C

Wash three times with PBS/Tween (200μΙ in the machine)

- Add 50 μΙ (each well) of secondary Ab (IgG Sigma, anti-mouse HRP) diluted 1 :16000 in Dilution buffer.

- Incubate plates 1 hr at 37°C

Wash three times with PBS/Tween (200μΙ in the machine)

Add to each well 50 μΙ of substrate Prestained TMB ONE Ready-to-use (KEM EN TEK)

Leave the plates on the dark starting from 15 min to 1 hr, preferably for 30 min When a blue colour has developed, stop the reaction by adding 50 μΙ of 1 N H ₂ S0 ₄ . The colour turns to yellow.

Read the plate by absorbance at 450 nm in the Victor ( Perkin Elmer)

- Positive values are considered at least two fold higher than blank.

Data were analyzed with ADAMSEL FPL, a free-software for non-commercial users.

Micro-neutralisation assays (MNA) Micro neutralisation assays (MNA) were carried out to assess VLP-immunised mouse sera for anti-influenza antibodies.

Firstly, virus stocks were titrated. A series of Iog10 dilutions were made and 0.1 ml/well (10 wells per dilution) was added into flat-bottomed 96-well plates containing a monolayer of confluent MDCK cells. Plates were incubated at room temperature for 30 minutes before replacing inoculum with infection medium (DMEM containing 2mM glutamine, sodium bicarbonate , penicillin-streptomycin 1/100, amphotericin B and 0.0025μg/ml TPCK trypsin). Plates were further incubated for 72 hours at 35°C. 50μΙ per well supernatants were harvested and run in HA assays using 0.7% turkey red blood cells. 50% Tissue culture infectious doses (TCID50) was calculated using the Spearman-Karber formula.

Sera samples were heat treated at 56°C for 50 minutes then added in duplicate into flat- bottomed 96-well plates using a starting dilution of 1/20, followed by a further seven doubling dilutions. 10 ² TCID50 (100μΙ) virus was then added into each well. Plates were incubated at room temperature for 1 hour before adding the mixtures to flat-bottomed 96-well plates containing a monolayer of confluent MDCK cells. After 30 minutes incubation at room temperature the serum-virus mixture was replaced with 100μΙ of infection media and incubated for 72 hours at 35°C. Supernatants were screened using an HA assay, as above. Serum neutralisation titres were expressed as the reciprocal of the highest dilution whereby 50% infection was prevented. Titres were the average of duplicate samples. Each assay run included a back -titration of the viruses used and validation criteria of 10 ² +/- 10° ⁵ /100μΙ. Values above threshold of detection in this assay of >2560 are reported as 5120; values below threshold of detection in this assay of <20 are reported as 10. F. Results

Six groups of five mice were immunised with five monovalent VLP or with a mixture of five VLPs on days 0, 28 and 56. Sera for analysis were collected on day 70. The total HA dose was constant at 1.5 thus the mix of five group received 0.3 μ9 HA per strain. Four groups of five mice were immunised on days 0, 28 and 56 with N1 VLPs, either monovalent or a mixture of three VLPs. Sera for analysis were collected on day 70. The total antigen (NA) dose was constant at 0.5 μ9.

In a second experiment six groups of 8 mice were immunised with monovalent and polyvalent VLPs as well as mixtures of H1 , H3 and B VLPs. Mice were immunised on days 0, 28 and 56 with 1 .5 μg of VLPs or 4.5 μg for the H1 , H3 and B mixture groups. Two groups of 8 mice were immunised on days 0, 28 and 56 with polyvalent N1 VLP expressing three N1 variants on each VLP or combined with H1 polyvalent VLP expressing three H1 antigens. The total antigen (NA) dose was constant at 0.5 μg. Blood samples for analysis were collected on day 70. In order to verify the immunogenicity of 0.3 pg versus 1 .5 pg HA an experiment was performed with A/Texas/36/1991 H1 VLPs. Mice immunised with the high dose had about twice the IgG levels (2.1 95% CI: 0.0 to 4.6) as mice immunised with the high dose. The microneutralisation (MN) and ELISA IgG results from both mouse studies are presented in Figures 1-6. Immune responses were evaluated using MN with cytopathic and Haemagglutination read out using the procedures as described in E. ELISA was performed using egg-grown whole influenza virus as coating antigen. ELISA IgG titres are expressed as Arbitrary Units (AU), where 1 AU/mL yields an OD value of 1 over blank. Thus a titre of 100 AU/mL indicates that a serum can be diluted 100 fold and yield an OD of 1 over blank.

H1 N1 strains

Table 1 and Figure 1 show the MN titres for H1 N1 strains. The data show that immunisation with a monovalent VLP only yields MN responses to the homologous antigen. Unexpectedly, the NC99 and Bris07 components show significant cross-reactivity in MN. The vaccines with multiple components (H1T and H1 P) do show broader reactivity, albeit at the cost of MN titre.

Table 1 : H1 N1 micro-neutralisation assay titres

Vaccine PR8 USSR77 Tex91 NC99 Bris07 Cal09

PR8 1832 (503 to 6663) 10 (10to 10) 10 (10to 10) 10 (10to 10) 10 (10to 10) 10 (10to 10)

USSR77 10 (10to 10) 263 (31 to 2239) 10 (10to 10) 10 (10to 10) 10 (10to 10) 10 (10to 10)

Tex91 10 (10to 10) 10 (10to 10) 43 (9to 206) 10 (10to 10) 10 (10to 10) 10 (10to 10)

NC99 54 (3 to 1097) 10 (10to 10) 10 (10to 10) 199 (38to 1046) 50 (14to 179) 10 (10to 10)

Bris07 10 (10to 10) 10 (10to 10) 10 (10to 10) 469 (205 to 1070) 25 (5 to 119) 10 (10to 10)

Mix 5 317 (76to 1324) 41 (23 to 72) 10 (10to 10) 10 (10to 10) 86 (41to 181) 10 (10to 10)

HIT 230 (140to 378) 10 (10to 10) 19 (10to 35) 44 (22 to 91) 16 (9 to 28) 10 (10to 10)

HIP 342 (237to 492) 32 (8to 125) 37 (17to 84) 115 (65 to 202) 25 (10to 63) 10 (10to 10)

PR8: monovalent VLP of HA variant A/Puerto Rico/8/1934; USSR77: monovalent VLP of HA variant A/USSR/92/1977; Tex91 : monovalent VLP of HA variant A/Texas/36/1991 ; NC99: monovalent VLP of HA variant A/New Caledonia/20/1999; Bris07: monovalent VLP of HA variant A/Brisbane/59/2007; Mix 5: mixture of all five monovalent VLPs (PR8+USSR77+Tex91 +NC99+Bris07); H1 T: polyvalent VLP of HA variants A/Puerto Rico/8/1934 + A/Texas/36/1991 + A/Brisbane/59/2007; H1 P: polyvalent VLP of HA variants A/Puerto Rico/8/1934 + A/USSR/92/1977 + A/Texas/36/1991 + A/New Caledonia/20/1999 + A Brisbane/59/2007; Cal09: A/California/07/2009

The results obtained with MNA are paralleled by the IgG data (Table 2 and Figure 2). Monovalent VLPs mainly induce homologous IgG responses. The cross reactivity between the NC99 and Bris07 components is also observed for the IgG response. The vaccines with multiple components (H1T and H1 P) again show broader reactivity, albeit at the cost of IgG titre.

Table 2: H1 N1 IgG titres

Vaccine PR34 USSR77 Tex91 NC99 Bris07 Cal09

PR8 5142 (975 to 27125) 2452 (852 to 7056) 2068 (624 to 6857) 3504(1036 to 11843) 1144 (531 to 2467) 1622 (882 to 2983)

USSR77 121 (31 to 482) 44737 (21485 to 93152) 4568(1299 to 16063) 4957 (2156 to 11394) 2076 (674 to 6387) 798 (82 to 7746)

TX91 277 (88 to 871) 2484 (659to 9358) 68661 (39105 to 120556) 8759 (4044 to 18971) 3366 (1254 to 9038) 725 (113 to 4662)

NC99 356 (96 to 1313) 3692 (1013 to 13457) 7811 (2523 to 24175) 71162 (23105 to 219175) 44994 (10710to 189030) 1482 (342 to 6419)

BR07 1060 (328to 3425) 8925 (2786 to 28593) 17594 (4325 to 71579) 106698 (37817 to 301042) 109132 (77909to 152867) 484 (86 to 2714)

MoMx 4641 (1897 to 11355) 13064 (8938 to 19094) 23450(13296 to 41357) 16465 (10618to 25532) 19822 (11352 to 34610) 1259 (672 to 2360)

TX91s 367 (152 to 885) 3699 (2005 to 6824) 78683 (59106 to 104745) 10856 (6809 to 17307) 3229 (1468to 7102) 1541 (935 to 2538)

Hm x 254 (83 to 773) 651 (330to 1285) 24976 (14950 to 41728) 6854 (4365 to 10763) 1345 (595 to 3039) 1323 (909to 1928)

HIT 22161 (15477to 31731) 2990 (778 to 11489) 46979 (36745 to 60062) 13137 (9029 to 19114) 21671 (13732 to 34200) 1308 (562 to 3047)

HITMx 13030 (9434 to 17997) 3222 (1475 to 7041) 31528 (20787 to 47819) 9260 (6387 to 13426) 12278 (8078 to 18661) 923 (442 to 1930)

HIP 24760 (18137 tD 33803) 8993 (5931 to 13635) 37688 (28890 to 49165) 15103 (9910 to 23017) 25085 (15331 to 41047) 904 (543 to 1505)

HIPMx 6501 (3972 to 10641) 3366 (1766 to 6416) 11214 (6882 to 18274) 10419 (8432 to 12874) 14950 (11370 to 18672) 419 (113 to 1554)

PR8: monovalent VLP of HA variant A/Puerto Rico/8/1934; USSR77: monovalent VLP of HA variant A/USSR/92/1977; Tex91 : monovalent VLP of HA variant A/Texas/36/1991 ; NC99: monovalent VLP of HA variant A/New Caledonia/20/1999; BrisOJ: monovalent VLP of HA variant A/Brisbane/59/2007;

Cal09: A/California/07/2009; MoMx: mixture of all five monovalent VLPs

(PR8+USSR77+Tex91 +NC99+Bris07); TX91s: monovalent VLP of HA variant ATexas/36/1991 ;

HmMx: mixture of monovalent VLPs of H1 (Tex91 ), H3 (SI87) and B (Yam88) subtypes (4.5ug total dose); H1 T: polyvalent VLP of HA variants A/Puerto Rico/8/1934 + A/Texas/36/1991 +

A/Brisbane/59/2007; HI TMx: mixture of polyvalent VLPs of H1 (H1 T), H3 (H3T) and B (HBT) subtypes (4.5ug total dose); H1 P: polyvalent VLP of HA variants A/Puerto Rico/8/1934 +

A/USSR/92/1977 + A/Texas/36/1991 + A/New Caledonia/20/1999 + A/Brisbane/59/2007; H I PMx:

mixture of polyvalent VLPs of H1 (H1 P), H3 (H3P) and B (HBP) subtypes (4.5ug total dose).

H3N2 strains

Table 3 and Figure 3 show the MN titres for H3N2 strains. The data show that immunisation with a monovalent VLP only yields MN responses to the homologous antigen. The vaccines with multiple components do show broader reactivity, albeit at the cost of MN titre. The only vaccine capable of inducing neutralisation titres to all vaccine components was the pentavalent H3 vaccine (H3P). The pentavalent H3 vaccine also shows broadening as reflected in the H3N2 MNA titres to 2013 (SW13) and 2014 (HK14) H3N2 variants. Table 3: H3N2 micro-neutralisation assay titres

Vaccine HK68 EN77 SI87 J094 FU03 HK14 SW13

HK68 nd 10 (10 to 10) 10 (10 to 10) 10 (10 to 10) 10 (10 to 10) 10 (10 to 10) nd

EN77 nd 671 (256 to 1757) 10 (10 to 10) 10 (10 to 10) 10 (10 to 10) 10 (10 to 10) nd

SI87 nd 10 (10 to 10) 11 (8 to 17) 47 (20 to 112) 10 (10 to 10) 10 (10 to 10) nd

J094 nd 10 (10 to 10) 10 (10 to 10) 485 (303 to 777) 10 (10 to 10) 10 (10 to 10) nd

FU03 nd 10 (10 to 10) 10 (10 to 10) 10 (10 to 10) 554 (265 to 1159) 10 (10 to 10) nd

Mix 5 nd 168 (40 to 699) 14 (8 to 27) 539 (284 to 1021) 299 (100 to 896) 10 (10 to 10) nd

H3T 263 (214 to 324) 22 (12 to 42) 17 (10 to 28) 30 (10 to 91) 1438 (667 to 3099) 10 (10 to 10) 10 (10 to 10)

H3P 165 (107 to 253) 153 (112 to 209) 20 (9 to 45) 392 (126 to 1224) 1182 (405 to 3445) 26 (12 to 56) 26 (12 to 56)

HK68: monovalent VLP of HA variant A/Hong Kong/1/1968; EN77: monovalent VLP of HA variant A/England/321/1977; SI87: monovalent VLP of HA variant A/Sichuan/2/1987; J094: monovalent VLP of HA variant A/Johannesburg/33/1994; FU03: monovalent VLP of HA variant A/Wyoming/3/2003; Mix 5: mixture of all five monovalent VLPs (HK68+EN77+SI87+JO94+FU03); H3T: polyvalent VLP of HA variants A/Hong Kong/1/1968 + A/Sichuan/2/1987 + A/Wyoming/3/2003; H3P: polyvalent VLP of HA variants A/Hong_Kong/1/1968 + A/England/321/1977 + A/Sichuan/2/1987 + A/Johannesburg/33/1994 + A/Wyoming/3/2003

The H3N2 IgG data show similar trends as MNA (Table 4 and Figure 4). Again monovalent vaccines mainly induce strain-specific IgG. The vaccines with multiple components (incl. the pentavalent H3 vaccine - H3P) show increased breadth.

Table 4: H3N2 IgG titres

Vaccine HK68 EN77 SI87 J094 FU03

HK68 1256 (459 to 3438) 4399 (788 to 24572) 1302 (850 to 1992) 7771 (3555 to 16984) 7303 (3032 to 17594)

EN77 308 (145 to 654) 156714 (77623 to 316392) 2604 (882 to 7690) 19327 (8024 to 46553) 9142 (3130 to 26700)

SI87 455 (110 to 1875) 36305 (14215 to 92726) 2391 (831 to 6877) 17792 (5960 to 53115) 11186 (4289 to 29175)

J094 198 (114 to 344) 7965 (4338 to 14623) 814 (284 to 2333) 55163 (37112 to 81994) 3571 (1865 to 6836)

FU03 389 (154 to 983) 12210 (3519 to 42368) 2459 (635 to 9519) 14001 (7449 to 26315) 25404 (11625 to 55515)

Mix 5 1442 (823 to 2528) 54914 (30914 to 97549) 1666 (752 to 3693) 44054 (32585 to 59558) 11676 (6779 to 20110)

5187s 1796 (1037 to 3107) 32842 (22394 to 48164) 2049 (1217 to 3449) 33167 (20182 to 54507) 11439 (7713 to 16965)

Mo x 1629 (703 to 3777) 19528 (14490 to 26319) 1775 (1035 to 3043) 14533 (11082 to 19059) 12771 (5953 to 27395)

H3T 11406 (8500 to 15306) 14957 (12095 to 18497) 1366 (914 to 2043) 9579 (6566 to 13974) 34428 (23993 to 49403)

H3TMx 6341 (4162 to 9661) 8186 (4587 to 14610) 1130 (580 to 2205) 4178 (2189 to 7973) 23232 (12998 to 41526)

H3P 9857 (5631 to 17255) 33331 (20848 to 53289) 1891 (886 to 4037) 21227 (10633 to 42375) 35456 (17591 to 71464)

H3PMx 11101 (6605 to 18658) 21427 (12667 to 36247) 1134 (810 to 1590) 15948 (10205 to 24922) 33765 (19894 to 57306)

HK68: monovalent VLP of HA variant A/Hong Kong/1/1968; EN77: monovalent VLP of HA variant A/England/321/1977; SI87: monovalent VLP of HA variant A/Sichuan/2/1987; J094: monovalent VLP of HA variant A Johannesburg/33/1994; FU03: monovalent VLP of HA variant A Wyoming/3/2003; Mix 5: mixture of all five monovalent VLPs (HK68+EN77+SI87+JO94+FU03); SI87s: monovalent VLP of HA variant A/Sichuan/2/1987 (bridging group); MoMx: mixture of monovalent VLPs of H1 (Tex91 ), H3 (SI87) and B (Yam88) subtypes (4.5ug total dose); H3T: polyvalent VLP of HA variants A/Hong Kong/1/1968 + A/Sichuan/2/1987 + A/Wyoming/3/2003; H3TMx: mixture of polyvalent VLPs of H1 (H1 T), H3 (H3T) and B (HBT) subtypes (4.5ug total dose); H3P: polyvalent VLP of HA variants A/Hong Kong/1/1968 + A/England/321/1977 + A/Sichuan/2/1987 + A/Johannesburg/33/1994 + A/Wyoming/3/2003; H3PMx: mixture of polyvalent VLPs of H1 (H1 P), H3 (H3P) and B (HBP) subtypes (4.5ug total dose). B strains

MNA responses following vaccination with monovalent or multiple component formulations again show a strong homologous response for the monovalent vaccines, whereas responses to the multicomponent compositions are broadened (Table 5 and Figure 5). The pentavalent formulation (HBP) again induced MNA titres to all vaccine component.

Moreover, the pentavalent formulation also induced MNA titres to recent B strain variants of both Victoria (Bris08) and Yamagata lineages (Phu13).

Table 5: B strain micro-neutralisation assay titres

Vaccine HK73 Vic87 Yam88 Jian03 Ma 104 Bris08 Phul3

HK73 82 (18 to 383) 10 (10 to 10) 10 (10 to 10) 10 (10 to 10) 10 (10 to 10) nd 10 (10 to 10)

Vic87 10 (10 to 10) 17 (8 to 36) 10 (10 to 10) 10 (10 to 10) 20 (8 to 47) nd 10 (10 to 10)

Yam88 10 (10 to 10) 10 (10 to 10) 54 (32 to 93) 10 (10 to 10) 10 (10 to 10) nd 22 (7 to 64)

Jian03 10 (10 to 10) 10 (10 to 10) 10 (10 to 10) 108 (63 to 185) 10 (10 to 10) nd 18 (6 to 50)

Mal04 10 (10 to 10) 34 (11 to 101) 10 (10 to 10) 10 (10 to 10) 76 (22 to 264) nd 10 (10 to 10)

Mix 5 80 (44 to 147) 54 (32 to 93) 15 (5 to 48) 71 (23 to 223) 94 (71 to 124) nd 42 (13 to 135)

HBT 67 (36 to 125) 10 (10 to 10) 10 (10 to 10) 26 (20 to 34) 16 (8 to 33) 14 (8 to 27) 14 (8 to 25)

HBP 43 (18 to 99) 41 (21 to 80) 26 (15 to 44) 13 (8 to 20) 98 (43 to 221) 28 (13 to 57) 26 (13 to 55)

HK73: monovalent VLP of HA variant B/Hong Kong/8/1973; Vic87: monovalent VLP of HA variant B/Victoria/02/1987; Yam88: monovalent VLP of HA variant B/Yamagata/16/1988; Jian03: monovalent VLP of HA variant B/Jiangsu710/2003; Mal04: monovalent VLP of HA variant B/Malaysia/2506/2004;

Bris08: B/Brisbane/60/2008, Vic; Phu13: B/Phuket/3073/2013, Yam; Mix 5: mixture of all five monovalent VLPs (HK73+Vic87+Yam88+Jian03+Mal04); HBT: polyvalent VLP of HA variants B/Hong Kong/8/1973 + B/Jiangsu710/2003 + B/Malaysia/2506/2004; HBP: polyvalent VLP of HA variants B/Hong Kong/8/1973 + B/Victoria/02/1987 + B/Yamagata/16/1988 + B/Jiangsu/10/2003 + B/Malaysia/2506/2004.

The B strain IgG data parallel the MN data (Table 6 and Figure 6). Monovalent vaccines only induce strain-specific IgG, whereas multi component vaccines induce broader responses at the cost of antibody levels. Again the pentavalent VLP vaccine (HBP) yielded the best results.

Table 6: B strain IgG titres

Vaccine BVic BYam BJian BMal BBris BPhu

HK73 24500 (11237 to 53419) 31945 (11731 to 86986) 276 (63 to 1214) 10020 (4232 to 23724) 3375 (202 to 56505) 1545 (248 to 9622)

Vic87 42408 (9830 to 182948) 26500 (11447 to 61346) 71 (28 to 185) 12656 (5348 to 29948) 417 (58 to 2988) 615 (243 to 1554)

Yam88 12097 (6155 to 23775) 29446 (14304 to 60620) 1183 (749 to 1868) 4810 (1950 to 11862) 489 (142 to 1681) 4985 (2094 to 11866)

Jian03 17817 (9717 to 32670) 20988 (16133 to 27304) 13989 (7314 to 26757) 10087 (5753 to 17686) 975 (247 to 3846) 8339 (4500 to 15453)

Mal0 61158 (23190 to 161285) 7048 (1353 to 36726) 223 (24 to 2108) 93599 (40985 to 213757) 7179 (2882 to 17878) 1094 (106 to 11275)

Mix 5 87443 (63484 to 120444) 30821 (16938 to 56082) 7332 (5414 to 9929) 73840 (50200 to 108613) 11403 (5714 to 22756) 7662 (5719 to 10264)

HB3s 8231 (5555 to 12197) 24644 (13479 to 45057) 852 (305 to 2382) 6396 (3988 to 10257) 480 (189 to 1218) 4855 (2770 to 8508)

MoMx 3854 (2321 to 6400) 10129 (6447 to 15915) 460 (195 to 1088) 3750 (1641 to 8569) 221 (87 to 564) 1601 (765 to 3350)

HBT 51281 (33450 to 78618) 14015 (9598 to 20464) 5747 (3789 to 8716) 97706 (54836 to 174088) 10372 (6469 to 16628) 3458 (1962 to 6094)

HBTMx 15307 (8502 to 27558) 4894 (2833 to 8454) 169 (523 to 5488) 47024 (30283 to 73019) 4442 (2454 to 8040) 722 (227 to 2299)

HBP 85288 (55132 to 131940) 20830 (13164 to 32959) 4439 (3352 to 5880) 151183 (111484 to 205018) 11949 (5929 to 24081) 4139 (2470 to 6935)

HBPMx 11420 (5036 to 25900) 2936 (1882 to 4581) 708 (461 to 1088) 37007 (24936 to 54922) 1638 (610 to 4395) 893 (425 to 1877) HK73: monovalent VLP of HA variant B/Hong Kong/8/1973; Vic87: monovalent VLP of HA variant B/Victoria/02/1987; Yam88: monovalent VLP of HA variant B/Yamagata/16/1988; Jian03: monovalent VLP of HA variant B/Jiangsu710/2003; Mal04: monovalent VLP of HA variant B/Malaysia/2506/2004; Bris08: B/Brisbane/60/2008, Vic; Phu13: B/Phuket/3073/2013, Yam; Mix 5: mixture of all five monovalent VLPs (HK73+Vic87+Yam88+Jian03+Mal04); HBT: polyvalent VLP of HA variants B/Hong Kong/8/1973 + B/Jiangsu710/2003 + B/Malaysia/2506/2004; HBP: polyvalent VLP of HA variants B/Hong Kong/8/1973 + BA ictoria/02/1987 + B/Yamagata/16/1988 + B/Jiangsu710/2003 + B/Malaysia/2506/2004. Summary

The data presented clearly indicate that broadening of anti-influenza responses occurs and requires polyvalent VLP with at least three, preferably at least five or even more components. For the H3 and B strains the response was broadened beyond the vaccine composition, i.e. strains isolated -10 years after the last strain included in the vaccine are neutralised by pentavalent VLP immunised mouse sera. The H3N2 and B strain data show that at least three, preferably five components on a single VLP induce broadened responses.

For H1 N1 , the mix of 5 as well as the pentavalent covered all vaccine components (i.e. MNA and IgG titres induced). A broadening could not yet be measured. It is assumed that the MN assay lacks sensitivity to detect A/California/7/2009 responses in mouse serum as has been previously observed with mini HA stem antigen constructs (Impagliazzo et al. (2015), Science 349(6254), 101 -106).