WO2011088451A1 - Uses of influenza virus-like particles (vlps) for characterization of neuraminidase and hemagglutinin activity - Google Patents

Abstract

The present invention provides methods of determining influenza antibody activity in a sample using an influenza VLP. The influenza VLP can be used to perform a hemagglutinin-inhibition (HAI) test or to perform a neuraminidase-inhibition (NI) test. The influenza VLP can also be used to quantify the activity of a small molecule inhibitor (SMI) of NA or other viral proteins.

Description

USES OF INFLUENZA VIRUS-LIKE PARTICLES (VLPs) FOR CHARACTERIZATION OF NEURAMINIDASE AND HEMAGGLUTININ ACTIVITY

CROSS REFERENCE TO RELATED APPLICATIONS

[001] This application claims the benefit of U.S. Provisional Application No. 61/295,434 filed on January 15, 2010 and U.S. Provisional Application No. 61/353,682 filed on June 1 1 , 2010, each of which is hereby incorporated by reference in its entirety for all purposes.

BACKGROUND

[002] The immunogenicity of influenza vaccines has historically been evaluated by measuring the antibody titer to hemagglutinin (HA), the major virus surface glycoprotein. Antibody to the the other major surface glycoprotein, neuraminidase (NA), essentially enhances protective immunity induced by influenza vaccines (Sylte and Suarez, 2010; Sylte et al. 2007). NA removes sialic acid from both viral and host proteins and participates in the release of viruses from infected cells. Only functional antibodies which inhibit the enzyme are protective and provide so-called permissive immunity: they do not prevent viral infection by themselves but significantly reduce virus spreading throughout the body and the severity of disease (Johansson et al., 1989; Sylte and Suarez, 2010).

[003] Although the concept of an influenza vaccine with HA and active NA present together has gained acceptance (Sylte and Suarez, 2010), no commercial vaccine with controlled NA activity is produced today. The lack of control for NA activity in influenza vaccines could be explained by a significant loss of enzyme activity during storage (Kendal et all., 1980). The problem with NA instability stands also behind the limited testing of anti-NA immune response in clinical studies.

[004] The principle of testing NAI inhibition by antibody includes the comparison of NA activity in samples incubated without and with NA-specific immune serum (Cate et al., 2010). There are several well-established methods for testing NA activity in viruses that are used for monitoring enzyme activity in vaccine during manufacturing (Kalbfuss et al., 2008) or screening mutants resistant to chemical NA inhibitors (Gubareva et al., 2002; Wetherall et al., 2003). For these applications the virus samples have highly variable NA activity as it is the object of analysis. In contrast, the neutralizing Ab assay requires a stable and reproducible source of active NA which needs to be qualified as any other analytical enzymatic reagent with a known absolute activity (enzyme units/ml) and storage life. Thus, existing publications discussing NA immunity are related rather to scientific proof of concept and have no indication of absolute activity and stability of NA.

[005] Using live influenza viruses for analytical purposes remains very complicated due to their pathogenicity, laborious and expensive preparation, and lack of standardization. Instead, inactivated virus vaccine (Cate et al., 2010) and purified NA protein (Johansson et all., 1989) were proposed as a more convenient source of neuraminidase. However, the conditions required for adequate measurement of NA inhibition with these substitutes of live virus, were not established. For vaccine preparations which most often have very low level of NA activity (Kendal et al.,1980), and high contribution of unfolded (denaturated) proteins (Feng et al., 2009), the requirements for NA qualification may include the enzyme stabilization over long-term storage and determining the range of NA activity providing comparable results with live virus. The active H1 N1 and H5N1 neuraminidases are now available from SinoBiological (Beijing, China) and RnD Systems (Minneapolis, USA). However, these preparations may contain unfolded NA molecules or a monomeric form of NA instead of a tetrameric complex present in virus (Sylte and Suarez, 2010) and their binding properties to antibodies could be modified.

[006] Recent progress achieved in creating influenza virus-like particles (VLPs) containing multiple viral proteins with relevant functional activity and immunoreactivity make VLP an excellent substitute for live viruses (Kang et al., 2009; Pushko et al., 2005; Lai et al., 2010). The unique combination of biological properties from original viruses, enhanced stability and ease in handling gives VLPs a big advantage over known virus substitutes i.e. inactivated virus vaccines and purified proteins. This results in expanding VLP applications for vaccine development and different virological studies (Kang et al., 2009; Lai et al., 2010). Unfortunately, NA in VLPs is not well characterized: only a few publications have reported detectable NA activity without its proper quantification in absolute units (Lai et al., 2010). Furthermore, it is still not clear if the activity and stability of NA in VLPs could be sufficient for analytical applications.

[007] The present application addresses the aforementioned problems associated with currently existing NA and HA activity assays by providing methods which use VLPs as a new source for active neuraminidase and hemagglutinin in NAI- and HAI-based antibody assays. SUMMARY OF THE INVENTION

[008] In a first aspect, the present invention provides methods of using influenza VLPs as substitutes for live virus for the characterization of HA and NA immune responses and for the screening of small molecule inhibitors. Such methods may be used to determine the influenza antibody activity in a sample.

[009] In one embodiment, the present invention provides a method determining influenza antibody activity in a sample, comprising: (a) incubating the sample with an influenza VLP comprising at least one hemagglutinin (HA) or neuraminidase (NA) moiety; and (b) measuring influenza HA or NA activity in the sample. In another embodiments, the antibodies in the sample neutralize HA or NA activity in the VLP. In yet another embodiment, the antibodies in the sample binds to the HA or NA.

[010] In one embodiment, the HA or NA is derived from a single subtype of influenza virus. In another embodiment, the VLP comprises HA or NA from multiple subtypes. In yet another embodiment, more than one VLP type is incubated with the sample.

[011] In one embodiment, the HA is derived from an influenza A, influenza B or influenza C virus. In another embodiment, the influenza virus is a pandemic influenza virus, such as H1 N1 or H5N1. In yet another embodiment, the HA subtype is selected from the group consisting of H1-H16.

[012] In one embodiment, the NA is derived from an influenza A, influenza B or influenza C virus. In some embodiments, the influenza virus is a pandemic influenza virus, such as H1 N1 or H5N1 . In some embodiments, the NA subtype is selected from the group consisting of N1-N9.

[013] In various embodiments described herein, the VLP may comprise a matrix protein. In one embodiment, the matrix protein is an influenza M1 protein. In another embodiment, the influenza M1 protein comprises a YKKL sequence (SEQ ID NO: 1 ) in the late domain.

[014] The HA activity in the sample can be measured by any suitable assay. In some embodiments, the HA activity comprises incubating the sample with red blood cells.

[015] The NA activity in the sample can be measured by any suitable assay. In some embodiments, the NA activity comprises incubating the sample with an NA substrate. In some embodiments, exposure of the substrate to NA results in release of a detectable reaction product. In some embodiments, the reaction product is fluorescent and/or luminescent. For example, In some embodiments, the NA substrate is 2'-(4- Methylumbelliferyl)-a-D-N-acetylneuraminic acid sodium salt hydrate (MUNANA) or a derivative thereof. In some embodiments, the reaction product is 4-Methyl Umbellipherone (MU). [016] In some embodiments, the methods of the present invention further comprise calculating an NA inhibition titer.

[017] The assay can be conducted by using a microplate of 96 wells. In some embodiments, the sample is incubated with a VLP in at least one well. In some embodiments, one VLP type is incubated with the sample in each well. In some embodiments, each VLP comprises a different NA subtype.

[018] In some embodiments, the sample is a serum sample. In some embodiments, the serum sample is a human serum sample. In some embodiments, the human has been vaccinated. In some embodiments, the vaccine is an inactivated influenza vaccine. In some embodiments, the vaccine is an influenza VLP vaccine. In some other embodiments, the vaccine is a live influenza vaccine.

[019] In some embodiments, the VLP retains at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, or more NA activity when stored for 6 months.

[020] As described herein, the present invention provides methods of using an influenza VLP comprising HA to evaluate the ability of an anti-HA antibody to block the binding of an influenza virus strain to red blood cells (RBCs). In some embodiments, said influenza virus is a pandemic influenza virus strain. In some embodiments, said RBCs are horse RBCs. In some embodiments, said method is performed at room temperature. In some embodiments, said method involves incubation of antisera with the antigen for at least about 3 hours.

[021] The present invention also provides methods of using an influenza VLP comprising neuraminidase (NA) to perform a neuraminidase-inhibition (Nl) test, said method comprising (a) incubating of said influenza VLP with and without anti-serum; and (b) measuring NA activity in the presence and absence of anti-serum to quantify NA inhibition by said antiserum.

[022] In another aspect, the present invention provides methods of using an influenza VLP for quantifying the activity of a small molecule inhibitor (SMI) of NA, said method comprising (a) incubating said influenza VLP with and without said SMI and (b) measuring NA activity in the presence and absence of said SMI to quantify NA inhibition by said SMI.

[023] In yet another aspect, the invention provides a novel RP-HPLC method to resolve and quantify intact HA proteins from influenza VLPs. This method is useful for determining the concentration of HA in influenza VLPs from any influenza strain and is capable of measuring HA concentrations in both homotypic and heterotypic VLPs. The results achieved using the methods of the present invention are in good agreement with concentrations determined by SRID. Importantly, the present RP-HPLC method is much quicker and less laborious and suitable for rapid quantification of HA in in-process and research samples. In various embodiments described herein, the novel RP-HPLC method can also be used to quantite additional proteins, including NA, M1 , and gp64.

[024] In yet another aspect, the present invention provides methods for the preservation of the enzyme activity of NA assembled on VLPs. The present invention describes an effective method to preserve the NA activity during BV inactivation of the VLP vaccine. The invention shows that by including Vitamin-E-TPGS and/or glycerol along with calcium in the inactivating buffer, the loss in NA activity for the influenza VLP vaccine during BPL exposure can be reduced significantly.

[025] The data presented herein shows that at concentrations greater than 0.05% w/v Vitamin-E-TPGS the NA activity is not only preserved at elevated temperatures (25°C) but actually enhanced compared to refrigerated conditions. This forms an additional part of the current invention. The invention describes the preservation of the NA activity through inclusion of Vitamin-E-TPGS in the final formulation of influenza VLP vaccine filled in final container/closer system and stored at temperatures greater than 25°C.

BRIEF DESCRIPTION OF THE FIGURES

[026] Figure 1 a represents typical chromatogram of H1 N1 VLP (HA concentration = 45.5 pg/ml) on RPHPLC. Figure 1 b represents typical chromatogram of H3N2 VLP (HA concentration = 75.5 pg/ml) on RPHPLC. Figure 1 c represents typical chromatogram of H1 N1 reference standards containing 9.4, 18.8, 37.5, 75, 150 pg/ml of HA. Figure 1 d represents typical standard calibration curve of HA in H1 N1 VLP (9.4, 18.8, 37.5, 75, 150 pg/ml of HA).

[027] Figure 2 represents peaks of M1 , NA, GP64, and HA in RP-HPLC and SDS-PAGE gel loaded with samples collected from HPLC, stained by coomassie-blue.

[028] Figure 3 represents chromatograms of blank, 25mM phosphate/0.5M NaCI, pH7.2, and 100 pg/ml HA A/Brisbane.

[029] Figure 4 represents chromatograms of 10, 50, 100, 150, and 200 pg/mL of HA (A/Brisbane H1 N1 ).

[030] Figure 5 represents % of HA and NA recovery on BPL treatment in presence of TPGS and various concentrations of calcium; Reaction buffer: 75 mM phosphate buffer, pH 8.0, 0.5M NaCI.

[031] Figure 6 represents % HA and NA recovery on BPL treatment in presence of glycerol and TPGS; Reaction buffer: 75 mM phosphate buffer, pH 8.0, 0.6M NaCI, 0.3 mM CaCI₂.

[032] Figure 7 represents NA activity in the presence of TPGS (7.5pg HA/ml target potency). Buffer A is phosphate buffer, pH 7.2 with 0.5 NaCI, 0.01 % w/v CaCI₂.2H₂0. [033] Figure 8 represents NA activity in the presence of TPGS (90pg HA ml target potency). Buffer A is phosphate buffer, pH 7.2 with 0.5 NaCI, 0.01 % w/v CaCI₂.2H₂0.

[034] Figure 9 represents calculation of log₂NIT using NA inhibition curves. NAI curves were plotted as Residual NA activity (NAA%) versus log₂[serum dilution] for preimmunized (log₂ (NIT1 ) and immunized (log₂ (NIT2) sheep serums using purified NA from B/Florida virus as an immunogen. NAI assay was performed with B/Florida VLPs (Lot 75508010). Solid lines indicate linear regression lines generated using part of inhibition curves around %Residual NAA = 75%. Dashed lines represent graphical determination of log₂NIT corresponded to 25% of NA inhibition for both curves. Calculated values for log₂ (NIT) were log₂ (NIT1 ) = 1 .72 and log₂ (NIT2) = 7.9.

[035] Figure 10 represents inhibition of different NA subtypes by NA(H5N1 A/lndo) - specific serum. NA inhibition curves in H5N1 A/lndo, H1 N1 A/NewCal, H3N2 A/NY and B/Sh VLPs by serum from sheep immunized with purified NA from H5N1 A/lndo virus (AB Titer 410,000).

[036] Figure 1 1 represents NA Inhibition by neat and prediluted serum samples with reduced anti-NA Ab level. Prediluted anti-H3N2 A Br ferret serum samples were prepared by dilutions of immune serum (NIT25 = 382.7) with preimmune serum (NIT25 =1.83) 1 :8 and 1 :256 and tested with H3N2 A/Brisbane VLP for NAI titer as per the assay procedure.

[037] Figure 12 represents Found log₂NIT versus expected log₂NIT. NAI titers were tested using H3N2 A/Brisbane VLP and anti- H3N2 A/Brisbane ferret serum samples prepared by dilution with preimmune serum. A dashed line indicates the line of identity and a solid line indicates the line of linear regression.

DETAILED DESCRIPTION OF INVENTION

Use of Influenza VLPs for Characterizing HA and NA Immune Responses:

[038] The influenza virus major surface glycoproteins HA and NA are the principal targets of the protective immune response. Licensed seasonal influenza virus vaccines are designed to elicit a protective immune response to the HA protein and are variable in their composition with regard to NA.

[039] As described herein, the hemagglutination inhibition (HAI) assay is the traditional standard assay used to evaluate the immunogenicity of influenza vaccines. Hemagglutination inhibition assays were introduced into medical and virology practice more than 60 years ago (Salk (1944) J. Immunol. 49, 87-98). Since that time, they have become important tools for measuring the efficacy of the anti-viral immunization, and for studying the neutralizing capacity of virus-specific antibodies. The protocol for HAI assays kept undergoing minor modifications (e.g., Cross (2002) Seminars in Avian and Exotic Pet Medicine 1 1 , 15-18; Hubby et al. (2007) Vaccine 25, 8180-8189; Wang et al. (2008) Vaccine 26 3626-3633; Rowe et al (1999) J Clinical Microbiology 37 937-943) to increase sensitivity and to try and make the assay more reproducible between loaboratories..

[040] In classical HAI assays, the antigen (e.g., live or inactivated virus), either as is, or pre-incubated with an antibody of interest, is mixed with a suspension of purified erythrocytes (red blood cells), e.g., human, dog, avian, equine, or murine erythrocytes, depending on the type of the viral antigen. After incubation of the mixture in V- or U- bottomed microwells, the visual effect can be two-fold: if the antibody is absent or unable to effectively block the virus, the latter links the erythrocytes into a dispersed, three- dimensional agglutinant; if the antigen is effectively blocked or absent, then the erythrocytes (ERCs) sediment to the bottom of the vial, forming the characteristic bright pellet, or "button."

[041] To determine the concentration or strength of a viral culture in hemagglutination assays, the sample is subjected to two-fold serial dilutions, until the agglutination vanishes. To determine the efficacy of the antiserum or tested antibody in the HAI assay, the sample is similarly subjected to serial dilution, until agglutination appears. The last dilution before the "borderline" between agglutination/non-agglutination is called the HAI titer.

[042] The present invention relates to the use of influenza virus-like particles (VLPs) as a substitute for live virus in order to characterize HA and NA immune responses and for screening small molecules.

[043] Virus-like particle (VLP) technology is capable of producing particulate biologic entities that have structural and immunological similarities to viral assemblies. Influenza VLP vaccines have been developed targeting both seasonal and pandemic strains of influenza virus. An example of an influenza VLP vaccine that has three proteins (hemagglutinin-HA, neuraminidase-NA and matrix-M1 ) captured within a lipid envelope secreted by Sf9 cells is described in US Patent No. 7,763,450, which is hereby incorporated by reference in its entirety for all purposes.

[044] Using VLPs instead of live virus for the characterization of HA and NA immune responses has several advantages. First, VLPs are not infectious and simplify requirements for assay setup especially for pandemic viruses which require enhanced biosafety containment. Second, VLPs provide highly active and stable neuraminidase in contrast to live viruses where NA activity is affected by different conditions and is vary variable. Having stable neuraminidase is especially important for conducting clinical trials which requires testing clinical samples in different sites over long period of time. Highly variable NA activity in live virus can affect the quality of data over the course of the assay. In addition, the maintenance of active NA in VLPs is much easier and cheaper than in live viruses. [045] Furthermore, like a live or inactivated influenza virus, VLPs have a quantifiable amount of functional HA protein on their surface, but unlike a live or inactivated virus, VLPs have no nucleic acids and are therefore not infectious. This is a significant advantage over the use of live virus, especially when dealing with BSL-3 viruses like H5N1.

[046] In addition, although the classical HAI assay is the accepted method for evaluating the activity of a seasonal influenza vaccine, it has not been proven to be a surrogate of protection when evaluating vaccines against pandemic influenza viruses such as H5N1 . A microneutralization (MN) assay, where Ab blocks actual virus infection of a target cell, has often been used in addition to or in place of the HAI assay. The present inventors and others have observed that in many instances (especially with pandemic strains of influenza), the HAI titer of a test antisera is lower than the corresponding MN titer. This has led researchers to wonder if there is a better HAI antigen that could be used in place of live virus to bring the HAI titer closer to that observed in the MN assay. The present invention addresses this need.

[047] In a first aspect, the present invention provides methods of using an influenza VLP comprising HA to evaluate the ability of an anti-HA antibody to inhibit an influenza virus. In some embodiments, the methods comprise a hemagglutination inhibition assay wherein the VLP is used as an antigen, replacing the role of a natural, live virus. In some embodiments, the methods comprise evaluating the ability of the anti-HA antibody to block the binding of the VLP to red blood cells (RBCs) in a hemagglutination inhibition assay. Compared to traditional methods of HAI assay using live influenza virus or purified HA antigen, the methods of the present invention using VLPs comprising the HA provide an HAI titer result which is more close to the titer result of a microneutralization (MN) assay.

[048] In some embodiments, the influenza VLP comprises one or more recombinant influenza virus matrix proteins, one or more recombinant influenza virus hemagglutinin (HA) proteins, and/or one or more recombinant influenza virus neuraminidase (NA) proteins. In various embodiments described herein, the HA and/or NA may exhibit hemagglutinin and/or neuraminidase activity, respectively. In additional embodiments, the HA and/or NA may be chimeric proteins. None-limiting exemplary methods of preparing VLPs are described in WO 2007/047831 , WO 2005/020889, US 7,763,450, and US 2010/0129401 , each of which is incorporated by reference herein in its entirety for all purposes.

[049] In some embodiments, the influenza virus is an avian, pandemic and/or seasonal influenza virus. The VLP can comprise HA or NA is from a single subtype of influenza virus, or HA or NA from multiple subtypes. In some further embodiments, the HA is selected from the group consisting of H1 , H2, H3, H4, H5, H6, H7, H8, H9, H10, H11 , H12, H13, H14, H15 and H16, and the NA is selected from the group consisting of N1 , N2, N3, N4, N5, N6, N7, N8 and N9. In some embodiments, the influenza virus is a pandemic virus, for example, the swine flu virus (novel H1 N1 ), or the bird flu virus (H5N1 ).

[050] The present invention also provides optimal conditions for conducting the HAI assays using VLPs as antigen. The assay can be selected from any immunology assays which are based on detecting inhibition of binding of HA to sialic acid receptors on cells or binding of HA to erythrocytes. The assay can be performed in microplates of 96 wells, or any other suitable containers depending on the setup.

[051] In some embodiments, the red blood cells are derived from a mammal species. In some embodiments, the mammal species is selected from the species in the Equidae family. For example, the species is a horse or horse-related species. In some embodiments, the concentration of red blood cells in the assay system is about 0.1 %, about 0.2%, about 0.3%, about 0.4%, about 0.5%, about 0.6%, about 0.7%, about 0.8%, about 0.9%, about 1.0%, about 1.1 %, about 1.2%, about 1 .3%, about 1.4%, about 1.5%, about 1.6%, about 1.7%, about 1.8%, about 1.9%, about 2.0%, about 3.0%, about 4.0%, about 5.0%, about 6.0%, about 7.0%, about 8.0%, about 9.0%, about 10.0%, or more.

[052] In some embodiments, the incubation time in the HAI assays using VLPs is about 1 hour, about 1.5 hour, about 2.0 hours, about 2.5 hours, about 3.0 hours, about 3.5 hours, or more.

[053] In some embodiments, the incubation temperature at room temperature, for example, about 16 °C, 17 °C, 18 °C, 19 °C, 20 °C, 21 °C, 22 °C, 23 °C, 24 °C, 25 °C, 26 °C, 27 °C, 28 °C, 29 °C, 30 °C, or more.

[054] In another aspect, the present invention provides methods of using an influenza VLP comprising NA to evaluate the ability of an anti-NA antibody to inhibit an influenza virus. In some embodiments, the methods comprise a neuraminidase inhibition assay wherein the VLP is used as an antigen, replacing the role of a natural, live virus. Similar to the HAI assay, the neuraminidase inhibition (NAI) assay is used as an investigative assay to evaluate the ability of influenza vaccines to elicit an immune response against neuraminidase. Typically it measures the titer of specific NA inhibiting antibodies (Ab) generated after immunization with NA-containing vaccine. The invention described herein substitutes VLPs for influenza virus as a source of active NA in the NAI assay.

[055] In some embodiments, the present invention provides methods of using VLP as a new source for active neuraminidase in NAI-based antibody assay. In some embodiments, to optimize the assay, the range for the initial NA activity and optimal NAI cut-off value for Ab titer estimation were established. In some embodiments, to set up an analytical range, the modified assay was validated for specificity, precision, accuracy, and linearity as per ICH guidelines (ICH, 1996). [056] In some embodiments, the methods comprise (a) incubating of said influenza VLP with and without anti-serum; and (b) measuring NA activity in the presence and absence of anti-serum to quantify NA inhibition by said anti-serum.

[057] To test NA activity, any fluorometric assay or a luminometric assay can be used, in some embodiments, fluorescence substrate in the assay of the present invention is fluorogenic 2'-(4-Methylumbelliferyl)-a-D-N-acetylneuraminic acid sodium salt hydrate (MUNANA) or functional equivalents. MUNANA is a substrate which gives comparable results for NA activity from multiple sources as it is based on quantification of the enzymatic reaction product (Potier et al., 1979; Wetherall et al., 2003 Kalbfuss et al., 2008).

[058] No MUNANA-based Ab assay has an established analytical range or has been validated previously. The present invention provides adjustments to the MUNANA-based NAI assay for testing Ab titer. For example, the present invention provides instructions to determine 1 ) the acceptable range of initial NA activity in VLP sample to provide consistent results for Ab titer; 2) whether VLP could be applied for testing Ab titer specific to only one subtype of NA in serum from subjects immunized with trivalent vaccine with no interference from Ab specific to HA and heterologous subtypes of NA; and 3) how the assay performance depends on the NAI cut-off value for Ab titer estimation and what is the optimal cut-off value.

[059] In addition, the present inventors have found that VLPs can be used as a substitute for live virus in screening for small molecule inhibitors of neuraminidase or any other viral proteins which could be incorporated in VLPs, one example being in drug discovery applications for anti-viral drugs. Accordingly, in yet another aspect, the present invention provides methods of using an influenza VLP for quantifying the activity of a small molecule inhibitor (SMI) of NA or other viral proteins. In some embodiments, the NA or other viral proteins are incorporated into the influenza VLP. In some embodiments, said method comprising (a) incubating said influenza VLP with and without said SMI and (b) measuring NA activity or other viral protein activity in the presence and absence of said SMI to quantify NA inhibition or other viral protein inhibition by said SMI.

[060] In any of the above-mentioned aspects and embodiments, proteins that can be incorporated in VLPs include, but are not limited to, influenza proteins (such as hemagglutinin (HA), neuraminidase (NA), matrix (M1 ), proton ion-channel protein (M2), nucleoprotein (NP), polymerase basic protein 1 (PB1 ), polymerase basic protein 2 (PB2), polymerase acidic protein (PA), and nonstructural protein 2 (NS2) proteins), or viral envelope proteins of other virus species.

[061] The invention also provides kits to perform the assays of the present invention. In some embodiments, the kits comprise one or more containers filled with one or more of the ingredients necessary for the assays. In one embodiment, the ingredients are packaged in a hermetically sealed container such as an ampoule or sachette indicating the quantity of composition. In one embodiment, the kits may further comprise reagents that can be used in a positive or negative control assay. The assay can be selected from any immunology assays which are based on detecting inhibition of binding of HA to sialic acid receptors on cells or binding of HA to erythrocytes. The assay can be performed in microplates of 96 wells, or any other suitable containers depending on the setup.

[062] Kits for using an influenza VLP comprising HA to evaluate the ability of an anti-HA antibody to inhibit an influenza virus are provided. In some embodiments, the kits comprise one or more ingredients comprising one or more suitable VLPs, depending on purpose. For example, the VLP comprises an HA protein from an influenza virus, wherein the HA can interact with the antibody to be tested. In some embodiments, the influenza virus is a pandemic influenza virus (e.g., H1 N1 or H5N1 virus). The kits can further comprise one or more ingredients and/or compartments that are necessary for an HAI asssay. For example, the kits comprise a solution containing red blood cells (RBCs). In some embodiments, the red blood cells are derived from an Equidae species.

[063] Kits for using an influenza VLP comprising NA to evaluate the ability of an anti-NA antibody to inhibit an influenza virus are provided. In some embodiments, the kits comprise one or more ingredients comprising one or more suitable VLPs, depending on purpose. For example, the VLP comprises an NA protein from an influenza virus, wherein the NA can interact with the antibody to be tested. In some embodiments, the influenza virus is a pandemic influenza virus (e.g., H1 N1 or H5N1 virus). The kits also comprise one or more ingredients and/or compartments that are necessary for an NAI asssay. For example, the kits comprise a fluorescent or luminescent substrate. In some embodiments, the fluorescence substrate is MUNANA, or functional derivatives.

The Development of an RP-HPLC Method for Quantitating VLP Protein Levels:

[064] As described herein, the concentration of HA protein has been standardized as a measure of potency of the vaccine across all manufacturers. Quantitation of the HA protein has traditionally been established using a single radial immunodiffusion (SRID) assay. However, the SRID assay is a laborious and lengthy test, making it difficult to provide rapid results for process development and research samples. Furthermore, developing SRID reagents for each new influenza vaccine strain is a rate limiting step in releasing vaccine.

[065] As described herein, a simple and reproducible reverse phase high performance liquid chromatography (RP-HPLC) method has been developed to quantitate proteins such as HA, NA, and M1 in VLP influenza vaccines. The method has been shown to accurately quantitate HA in numerous influenza strains of VLPs for in-process and research samples. In addition, VLP proteins such as NA, M1 , and baculovirus contaminants such as gp64 can also be resolved and identified in the same chromatogram together with HA, thus making it possible to quantitate all major VLP proteins in a single run.

[066] Thus, in another aspect, this invention provides a novel RP-HPLC method to resolve and quantify intact proteins, such as HA, from influenza VLPs. This method has been used to determine HA concentration in influenza VLPs from different strains. The results are in good agreement with the concentration determined by SRID. Moreover, the RP-HPLC method is much quicker and less laborious and suitable for rapid quantification of HA in in- process samples. The method can be used for quantification of HA during in-process and final bulk and vialed product of influenza VLPs.

[067] In one embodiment, the RP-HPLC method can be extended to determine the HA content in all strains of influenza VLPs. A further refinement may allow one to determine the concentration of HA in a new VLP based on a previous strain reference sample, e.g. 201 1 strains might be tested against 2010 reference samples before the corresponding reference samples are available. As described above, this method may also be extended to determine NA, M, and BV/SF9 protein content in VLP samples.

Methods of Preserving NA Activity of VLPs:

[068] As described above, the influenza virus major surface glycoproteins hemagglutinin (HA) and neuraminidase (NA) are the principal targets of the protective immune response. Licensed seasonal influenza virus vaccines are designed to elicit a protective immune response to the HA and NA proteins. However, only the concentration of HA protein is standardized in the currently approved inactivated seasonal influenza virus vaccines; the concentration of the NA protein is not. Hemagglutinin induces strain-specific neutralizing antibodies that prevent infection by antigenically related influenza viruses. Unlike HA-specific antibodies, NA-specific antibodies do not prevent influenza virus infection, and NA immunity is referred to as infection permissive. However, humoral immunity induced by NA can markedly reduce virus replication and release, shortening the severity and duration of illness, a reasonable goal in the event of an influenza pandemic.

[069] NA activity is needed for the release of progeny virus from infected cells and possibly also for penetration of the sialic acid-rich mucus layer lining the respiratory epithelia in order to reach the host cell. Blocking the activity of NA leads to arrest of viral infection and this idea has motivated the design of specific inhibitors based on the atomic structure of NA. NA is very sensitive to processing conditions. Hence, the majority of the commercial influenza vaccines have very limited quantity of "active" NA in them. Considering the importance of NA on overall performance of the influenza vaccine, it is desirable to retain its biological activity during manufacturing as well as in final product. Because NA is an enzyme, it is prone to inactivation under non-refrigerated conditions, especially at temperatures greater than room temperature.

[070] The NA of different strains of influenza virus varies in their stability at 37 °C. The enzymes of the strains with 1 neuraminidases were found to be unstable during incubation at 37 °C whereas the enzymes of the strains with the N2 neuraminidases were stable. The VLP technology uses genetically modified baculovirus (BV - designed to express therapeutic proteins - HA, NA and M1 ) to infect the insect cells (sf9 cells). At the time of harvest, there exists a population of BV along with VLPs. Because of the similarities in size and surface properties, these BV particles copurify with VLPs during downstream processing. It is necessary to inactivate the BV before the vaccine can be released for human use. Historically researchers have used heat, extreme pH, radiation, detergent / solvent mixtures, formalin and beta-propiolactone (BPL).

[071] On commercial scale, BPL treatment has proven to be a viable and very effective tool to clear the BV as it offers multiple advantages over other methods. However, when a mixture of BV and VLP is exposed to BPL, in addition to BV clearance, the BPL irreversibly damages the therapeutic proteins - HA, NA and M1. For a given set of exposure condition (temperature, BPL concentration and exposure time), damage to the NA is far greater compared to the other two proteins. For the influenza VLPs produced using present technology, there are two major conditions for NA inactivation: (i) viral clearance step during down-stream processing using beta-propiolactone (BPL), and (ii) storage of the final product, especially at non-refrigerated temperatures. Notably, during the BPL treatment step (meant to inactivate the baculovirus), more than 90% of the active NA gets destroyed. The present inventors solve this problem by providing methods for the preservation of the enzyme activity of NA assembled on VLPs.

[072] As described herein, NA activity can be preserved via the inclusion of Vitamin E TPGS and/or glycerol in addition to Ca²⁺ during treatment with BPL. Such a preservation method has not previously been described.

[073] The data demonstrated herein suggests that a relationship between Vitamin E TPGS and/or glycerol, calcium concentration, temperature, pH and NA activity exists for the influenza VLP vaccine. In particular, the data shows that including Vitamin E TPGS and/or glycerol along with calcium not only facilitates preservation of NA activity for the VLPs at elevated temperatures, but also enhances of the enzymatic activity at physiological temperatures (37°C). The effect of Vitamin-E-TPGS and/or glycerol on preservation of NA activity is also expected to prevail for other methods for BV inactivation (gamma irradiation, UV exposure, etc.). Therefore, this invention is also useful in formulations containing influenza VLPs with preserved NA activity at non-refrigerated conditions.

[074] This invention is further illustrated by the following examples that should not be construed as limiting.

EXAMPLES

Example 1 : Quantification of HA in Virus-Like Particles (VLPs)

[075] Method Outline: Preparation of sample and VLP reference standards: Add 1 1 pi of 10% zwittergent into 99 pL of sample or VLP reference standard and mix well.

[076] Preparation of mobile phase: Mobile Phase A: 0.1 % TFA in HPLC water; Mobile Phase B: 0.2% TFA in Acetonitrile

[077] Preparation of 1 % NP9 for column wash: Add 10 μΙ of NP9 into 990 μΙ of HPLC water.

[078] The HPLC method parameters are as follows:

HPLC method parameters for sample testing:

Column: Varian, PLRP S 4000A, 4.6x150mm

Column temp: 75 °C

DAD: 280 nm

Mobile Phase Gradient:

Time (min) Mobile Phase A (%) Mobile Phase B (%)

0.0 80 20

15 0 100

20 0 100

Post run: 2 min

Injection volume: 100 μΙ

The HPLC method parameter for column wash

Column temp: 75 °C

DAD: 280 nm

Mobile Phase Gradient:

Time (min) Mobile Phase A (%) Mobile Phase B (%)

0.0 80 20

7.5 0 100

10 0 100

Post run: 2 min

Injection volume: 10 μΙ of 1 % NP9 Data analysis

[079] Concentration of HA in sample is determined using linear regression of VLP reference standards. Typical chromatograms of H1 N1 VLP (HA concentration = 45.5 pg/ml), H3N2 VLP (HA concentration = 75.5 Mg/ml), and H1 N1 VLP reference standards (HA concentrations = 9.4, 18.8, 37.5, 75, and 150 Mg/ml) on RPHPLC are shown from Figure 1a, 1 b, and 1 c, respectively.

[080] A typical standard calibration curve of HA in H1 N1 VLP (HA concentrations = 9.4, 18.8, 37.5, 75, and 150 Mg/ml) is shown in Figure 1 d.

[081] Comparison of HA concentration determined by RPHPL and SRID in H1 N1 VLP samples is displayed in the table below:

Table 1. Comparison of Microneut Titer to HAI tiers with VLP and Virus Ag Using Sera from an H5N1 Clinical Trial

Example 2: Rapid Quantitation of HA in VLPs by RP-HPLC

[082] The following parameters were used to measure HA levels in influenza VLPs by RPHPLC: Mobile phase A: 0.1 % TFA in water; Mobile phase B: 0.1 % TFA in acetonitrile; Column: PLRP-S 4000A, 2.1X150mm, 8 Mm; Column temperature: 75 °C; Flow rate: 0.6 ml/min; Detector: DAD 280nm; Mobile phase gradient: Ramp mobile phase B from 30% at time 0 to 100% at time 15 min, keep 100% of B for 3 min.

[083] Sample Preparation: 11 pL of 10% zwittergent is added into 99 pL of sample to make

1 % zwittergent in sample and mix well with vortex mixer.

[084] Results: The peaks were identified as shown in Figure 2.

[085] The assay showed no interference from the system and buffer on the determination of HA (A/Brisbane H1 N1 ), see Figure 3.

[086] The correlation coefficient of 10, 50, 100, 150, and 200 pg/mL of HA (A/Brisbane H1 N1 ) standards was determined to be 0.99995, see Figure 4. [087] Six replicates of 100 g/ml HA (A/Brisbane H1 N1 ) were injected. % CV was determined to be 0.7, and % recovery was determined to be 98.0:

Table 2. HA Quantitation in VLPs

[088] Three different levels of HA (H1 N1 A/Brisbane) standards were injected in duplicate and accuracy is shown as follows:

Table 3. Accuracy of HA Standards

[090] HA (A/Brisbane H3N2) concentration can be estimated using HA (A/Brisbane H1 N1 ), and thus determination of HA (A/Brisbane H3N2) quantity can be made using the heterologous HA calibration curve: Table 5. HA Concentration Estimates

[091] Three different levels of HA (B/Brisbane) standards were injected in duplicaion and accuracy is shown as follows:

Table 6. Accuracy of HA Standards

[092] HA (B/Florida) concentration can be estimated using HA (B/Brisbane), and thus determination of HA (B/Florida) quantity can be made using the heterologous HA calibration curve (B/Brisbane):

Table 7. HA (B/Florida) Concentration Estimates

[093] A comparison of HA (A/Brisbane H1 N1 ) by RP-HPLC vs. SRID for in-process samples shows:

Table 8. Comparison of RP-HPLC vs. SRID for In-Process Samples

[094] A comparison of HA (A/Brisbane H1 N1 ) by RP-HPLC vs. SRID for final bulk samples shows: Table 9. Comparison of RP-HPLC vs. SRID for In-Process Samples

[095] For the in-process samples, HA concentrations by RP-HPLC were often greater than that determined by SRID because RP-HPLC determines chemical concentration and SRID determines biological concentration (activity) of HA. For the final bulk samples, RP-HPLC and SRID are in close agreement.

[096] These data show that this RP-HPLC method is capable of detecting and quantitating HA in influenza VLPs from various strains showing good specificity, linearity, precision and accuracy.

[097] For in-process samples, the HA concentration analyzed by RP-HPLC, is often greater than the SRID results. This trend is possibly attributed to the HPLC method detecting all HA protein present in the sample, regardless of retention of antigenicity. In contrast, the SRID will only detect and quantitate antigenic HA specific to the antibody present in the agarose gel. Analysis on purified final bulk material shows that the RP-HPLC results are very comparable to the HA concentration quantitated by the SRID method.

Conclusions:

[098] All HPLC peaks were identified using a fraction collector followed by SDS PAGE and Western blot analysis. One major challenge for in-process sample testing is the column pressure build-up due to the complex sample matrix. Influenza VLPs are composed of lipids and other hydrophobic components which can adhere tightly to the reverse phase column. Prolonged column exposure to these hydrophobic complexes will tend to deteriorate the column leading to poor performance. This condition can be avoided simply by filtering in- process samples through 0.22 pm filter before treatment with zwittergent detergent. An inline filter was tried but this method resulted in a reduced signal of the HA peak. An efficient column cleaning step was developed to clean the column in-between each sample injection by alternating the column flushing using a base (from 0.1 N to 0.5N NaOH), an acid (from 0.1 N to 0.5N HCI), and then the mobile phase B (0.1 % TFA in acetonitrile) through the column. This RP-HPLC cleaning method was found to be efficient in recovering the column. The data shows that the RP-HPLC method is also capable in quantitating both the NA and M1 proteins associated with the VLP. Additionally, BV contaminants (e.g. gp64) can also been quantitated via RP-HPLC. [099] One major challenge associated with using the RP-HPLC method is the column cleaning step after each injection of VLP. Since VLP is a large structure most reverse phase columns are not suitable for separation of proteins from an entity of this size. Due to the particle size, column pressure buildup is a common problem during this assay and development of an appropriate column cleaning step was necessary to remove column clogging material. Thus, an efficient column cleaning method was developed to recover the column after each injection of the VLP sample.

[0100] In addition, due to the complex matrix of the influenza VLP, upstream samples contain many hydrophobic components which bind tightly to the reverse phase column. These hydrophobic interactions result in a difficult eluting step which causes high column pressure and reducing peak resolution. PLRP S4000 reverse phase column was found to be appropriate for this use.

[0101] Proper disassociation of VLP is critical to release HA completely. Different detergent treatments were evaluated and 1 % zwittergent was selected as optimum condition. No reducing reagent or trypsin is needed in this method unlike previous RP-HPLC methods.

Example 3: Measurement of HAI Titers Using VLPs

[0102] To determine the utility of a VLP-based HAI assay, a series of experiments were conducted comparing HAI titers using VLP or live H5N1 virus as the Ag and comparing it to titers obtained in the MN assay.

[0103] Method Outline (Parameters Investigated):

a. Incubation Time of Antisera With Antigen

i. 60 mins at room temp in PBS (assay standard)

ii. 120 or 180 mins at room temp in PBS

b. Antigen Type

i. Live Indonesia H5N1 virus (assay standard)

ii. H5N1 VLP

c. RBC type

i. 1 % horse (assay standard)

ii. 0.5% turkey RBCs

iii. 0.75% human RBCs

iv. 0.75% guinea pig RBCs

d. Incubation Temp

i. Room Temp (assay standard)

ii. 37°C Table 10. Comparison of Microneut Titer to HAI Titers with VLP and Virus Ag Using Sera from an H5N1 Clinical trial (ND = Not Done)

[0104] Summary of the Results:

[0105] 4 experiments were conducted and 77 clinical samples were evaluated from subjects vaccinated with an H5N1 VLP vaccine.

[0106] HAI titers using VLP or Virus Ag were similar to each other, although the titer using VLP was consistently about 2-fold higher than that derived from using whole virus as Ag.

[0107] In every assay the HAI titer using VLP Ag was closer to the MN titer than the HAI titer using jnfluenza virus Ag.

[0108] Conditions identified as sub-optimal for conducting HAI assay included RBC type (Use of turkey, human or guinea pig RBC resulted in low HAI titers), and incubation temperature (Incubation at 37 °C resulted in low HAI titers).

[0109] Conditions identified as optimal for conducting HAI assay included incubation time (3 hr, as opposed to the standard 1 hr), antigen type (VLPs - the resulting value was closest to MN titer), RBC type (1% horse RBCs), and incubation temperature at room temperature.

Example 4: Measurement of NAI Titers Using VLPs

[0110] To determine the utility of a VLP-based NAI assay, a series of experiments were conducted comparing NAI titers using VLPs or live B/Florida virus as the source of active neuraminidase.

[0111] A fluorometric assay for testing NA activity in this study used MUNANA as a substrate. However, any other fluorometric or luminometric assay for testing NA activity could be used. [0112] Method Outline:

a. Incubation Time of Antisera with Antigen

i. 30 mins at room temp with continuous shaking

b. Antigen type (Active NA)

i. Live B/Florida virus

ii. B/Florida VLPs

c. Incubation with Substrate

i. 100 μΜ MUNANA

ii. 40 mins at 37°C with continuous shaking

d. Stop Reaction with Stop Reagent

i. 0.1 M glycine with 25% ethanol, ph 10.7

e. Fluorescence Measurement: Ex: 365 nm, Em: 420nm

f. The quantitation of NA activity was performed using calibration curve for MU and expressed in nmol MU/well

g. The % of NA activity relative to the blank control without antiserum (residual NAA) was plotted against Log2 antiserum dilutions by linear regression h. Log2 dilution of the sera where NA activity is reduced to 75% of control NAA (no sera) was defined as Log2 (NIT25) of antiserum. NA inhibition titers (NIT25) was reported as serum dilution back calculated from log2 dilution value.

Table 11. NA Inhibition Titers against B/FI VLP and B/FI live virus in serums immunized with VLP and TIV vaccines.

[0113] Summary of the Results:

[0114] The immune response against NA was tested using B/FI VLP and B/FI live virus in NAI assay for four groups of patients and evaluated by GMT (Geometric Mean) for NIT25, Post-to Pre Ratio (PPR) of NIT25 and % of patients with PPR > 4.0 For each parameter similar results with insignificant difference between VLP and live virus were obtained.

Example 5: VLPs as a substitute for live virus for screening small molecule inhibitors of NA

[0115] This example describes a fluorometric assay for testing NA activity using MUNANA as a substrate. However, any other fluorometric or luminometric assay for testing NA activity could be used.

[0116] Method outline is described as below:

a. Incubation time of NA inhibitor (Tamiflu) with antigen

i. 30 mins at room temp with continuous shaking

b. Antigen Type (active NA)

i. Live B/Florida virus, B/Florida VLPs, H3N2 VLPs, H1 N1 VLPs, or H5N1 VLPs

c. Incubation with substrate

i. 100 μΜ MUNANA

ii. 40 mins at 37°C with continuous shaking

d. Stop Rxn with Stop Reagent

i. 0.1 M glycine with 25% ethanol, ph 10.7

e. Fluorescence Measurement: Ex: 365 nm, Em: 420nm

f. The quantitation of NA activity was performed using calibration curve for MU and expressed in nmol MU/well

g. The % of NA activity relative to the blank control without Tamiflu (residual NAA) was plotted against Log2 [inhibitor] by linear regression h. Log2 [inhibitor] where NA activity is reduced to 50% of the control NAA (no sera) was defined as Log2 [IC50] of Tamiflu. 50% inhibition concentration (IC50) was reported as an inhibitor concentration back calculated from log2 [IC50] value.

Table 12. NA Inhibition (IC50) by Tamiflu (Oseltamivir Phosphate, Roche) in Influenza VLP and Live Viruses tested by NAI assay (IIC50 range provided by manufacturer from 0.8 nM to > 35 μΜ)

Sample Description IC50 μΜ

VLPs

H3N2 A Brisbane 0.14

H1 N1 A/Cal VLP 0.33

H5N1 VLP BPL090409 1 .32

B/Florida 5.19

Live Virus B/Florida 4/06/MD-4 32°C Crude SRNT 2.10

[0117] Summary of Results:

[0118] IC50 values obtained using VLPs or Live Virus were within the expected range provided by manufacturer. Two fold difference of IC50 values obtained with VLPs or Live Virus was not significant for such inhibition testing and was within the expected method accuracy ( two fold dilutions of inhibitor)

Example 6: Preservation of NA during BPL Inactivation Step:

[0119] Experiments were conducted to demonstrate the effect of TPGS on NA preservation (Figure 5 - showing % of HA and NA recovery on BPL treatment in presence of TPGS and various concentrations of calcium; Reaction buffer: 75 mM phosphate buffer, pH 8.0, 0.5M NaCI). In presence of 10μΜ calcium, the NA activity was lost almost completely after BPL treatment (0.2%). Inclusion of calcium at a concentration of 0.3 mM offered benefit in retention of NA activity. However with the inclusion of 0.05% w/v TPGS the retention of NA activity was increased further, particularly for the reaction buffer containing 0.3mM calcium. The HA potency was maintained within acceptable limits under all treatment conditions.

[0120] A second experiment was conducted, wherein the calcium concentration in reaction buffer was fixed at 0.3mM and the effect of addition of TPGS and/or glycerol during the BPL treatment was studied (Figure 6 - % HA and NA recovery on BPL treatment in presence of glycerol and TPGS; Reaction buffer: 75 mM phosphate buffer, pH 8.0, 0.6M NaCI, 0.3 mM CaCI₂). NA activity was retained best by using a combination of glycerol and TPGS during the BPL treatment step.

Example 7: Inactivation of NA upon Exposure to Non-Refrigerated Temperatures

[0121] The below table 13 shows a temperature-dependent degradation of NA activity for H5N1 Influenza VLP vaccine for all four potencies. None of these formulation contained calcium. At refrigerated storage condition, the NA activity remained stable as a function of time, but at 25°C, the NA activity declined gradually. The formulation included phosphate buffer (pH 7.2) with 0.5 M NaCI. The product was filled in 3 mL USP Type I glass vials.

Table 13. Temperature-Dependent Degradation of NA Activity (in terms of mU/mg TP)

90 pg HA/mL (5±3°C Upright Orientation)

67.9 48.6 66.2 64.0

90 pg HA/mL (25±3°C Upright Orientation)

67.9 25.9 7.7 2.5

30 pg HA/mL (5±3°C Upright Orientation)

33.8 36.8 48.8 28.3

30 pg HA/mL (25±3°C Upright Orientation)

33.8 13.8 4.8 1 .7

15 pg HA/mL (5±3°C Upright Orientation)

3.66 17.0 12.3 3.3

15 pg HA/mL (25±3°C Upright Orientation)

3.66 5.7 1.3 0

[0122] The data presented in the above table is in agreement with the literature - that NA activity is reduced at non-refrigerated temperatures.

[0123] An experiment was conducted to investigate the effect of adding TPGS on stability of NA present in H5N1 Influenza VLP vaccine, upon storage. Table 14 below shows formulations of standard and test. The fill volume was 0.6 mL per vial, with 2 vials per time- point per formulation. The primary packing material was 2 mL USP Type I vials and 13 mm grey chlorobutyl FluroTech stopper, with 13 mm flip-off aluminum seals. The storage conditions were 2 to 8°C and 25°C in upright orientations. The time points for analysis were 0, 1 , and 3 months.

Table 14. Formlations of Standard and Test.

[0124] Figure 7 shows NA activity in the presence of TPGS (7.5pg HA/ml target potency).

Buffer A is phosphate buffer, pH 7.2 with 0.5 NaCI, 0.01 % w/v CaCI₂.2H₂0.

[0125] Figure 8 shows NA activity in the presence of TPGS (90pg HA/ml target potency).

Buffer A is phosphate buffer, pH 7.2 with 0.5 NaCI, 0.01% w/v CaCI₂.2H₂0.

[0126] VLP samples with different formulations were stored at 2-8 and 25°C and tested after

1 month and 3 months. NA activity of samples formulated with 0.05% w/v TPGS was found to be better preserved even at non-refrigerated temperature over 3 months. This was particularly true for low dose samples (showing target potency 7.5pg HA/ml).

Example 8. Validation of NA antibody assay

Materials and methods

1. VLP preparation [0127] Purified VLPs are comprised of recombinant influenza virus hemagglutinin (HA), neuraminidase (NA) and matrix 1 (M1 ) proteins. Appropriate genes are cloned into a baculovirus expression vector, and VLPs are assembled from recombinant HA, NA, and M1 expressed in Spodoptera frugiperda (sf9) insect cells (ATCC CRL-1711 ) (Invitrogen, Carlsbad, USA). Abbreviations for influenza viruses or proteins used in this study are A/lndo (A/Indonesia 05/2005 (H5N1 )), A/Cal (A/California 04/2009 (H1 N1 )), A/NewCal (A/New Caledonia/20/99(H1 N1 )), H1 N1 A/Br (A/Brisbane 59/2007 (H1 N1 )), A/NY (A/New York/55/2004 (H3N2)), H3N2 A/Br (A Brisbane 10/2007 (H3N2)), B/Br (B/Brisbane 60/2008), B/FI (B/Florida 04/2006) and B/Sh (B/Shanghai/361/2002).

[0128] The cloning of influenza HA, NA, and M1 into baculovirus (BV) expression vectors was described previously (Pushko et al., 2005; Bright et al., 2007; Mahmood et al., 2008). Sf9 insect cells were infected with recombinant baculovirus. VLPs were purified from the supernatant harvested from sf9 cell culture (Bright et al., 2007). After removal of sf9 cells VLP preparation was concentrated and diafiltered through tangential flow filtration (TFF). Separation of VLPs from BV particles and contaminating DNA, RNA and sf9 proteins was achieved by anion exchange (IEX) and size-exclusion (SEC) chromatography. The VLP preparation was finally sterile filtered through a 0.22pm PVDF membrane and stored refrigerated (2-8 °C).

[0129] The expression of HA, NA, and M1 proteins in VLP preparations was confirmed by SDS-PAGE and Western blot analysis and HA concentration was quantified by a single- radial-immunodiffusion (SRID) as described (Pushko et al., 2005; Mahmood et al., 2008). Total protein (TP) concentration was tested using BCA assay (Pierce, Rockford, USA).

2. Purified neuraminidase and hemagglutinin preparations

[0130] NA and HA from H5N1 A lndo virus and NA from B/FI virus were purified by the following procedure. Corresponding native HA and modified NA (fusion of NA with HA secretion sequence) were individually cloned into the same recombinant BV expression vector used to produce VLP's. The Sf9 cells were infected and the proteins were expressed in a similar manner as VLP's described in this paper. The HA proteins were extracted from the cell paste and purified through TMAE anion exchange, lentil lectin affinity, and hydroxyapatite (CHT) chromatography. The secreted NA proteins were harvested from cell culture and purified through three column chromatography steps including Fractogel EMD TMAE, lentil lectin, and hydroxyapatite columns. Purified HA and NA proteins were > 95% pure as determined by SDS-PAGE, coomassie staining and densitometric scanning and were used for animal immunization.

3. Antisera against purified neuraminidase and hemagglutinin [0131] Sheep immunization with purified HA and NA antigens from B/FI, H3N2 A/NY and H5N1 virus strains, followed by blood collection and antiserum preparation were performed by Covance (Princeton, USA). Sheep were immunized ID/IM with 50 pg antigen in complete Freund's adjuvant, boosted three weeks later by the same route with 30 pg antigen mixed with in-complete Freund's adjuvant and every week thereafter. Antibody titers were determined by ELISA as described below.

4. Ferret antisera after immunization with A/Brisbane/10/07(H3N2) virus

[0132] Male ferrets (Mustela putorious furo) 3-4 months of age with confirmed seronegativity against H3N2 A Bris virus were used for immunization (Bioqual Inc, Rockville, USA). Egg grown H3N2 A/Bris virus was obtained as described (Bright et al., 2008) and given intranasally to ferrets ( n = 4) in a volume of 0.5 ml (250 μΙ per nostril), under anesthesia. Each ferret was inoculated with a total of 1 x 108 pfu/0.5 ml of virus. Twenty days after inoculation, blood was collected from the anesthetized ferrets via the anterior vena cava. Serum was separated and frozen at -80 ± 5oC. Ferrets were euthanized and exsanguinated on Day 35. All procedures were in accordance with the NRC Guide for the Care and Use of Laboratory Animals.

5. Human antisera after vaccination with trivalent VLP vaccine

[0133] Seasonal trivalent VLP vaccine comprised of H1 N1 A/Bris, H3N2 A/Bris, and B/FI VLP strains recommended for the 2008-2009 influenza vaccine was used in a clinical trial (manuscript in preparation). Normal healthy adults (18-50 years old) were randomized to receive VLP vaccine at 15 pg (85 subjects), 60 pg (85 subjects) or placebo (50 subjects). Blood samples were collected before (day 1 ) and after (day 22) immunization followed by serum preparation.

6. Antibody titer ELISA assay

[0134] Antibody (total IgG) titers in anti-HA and anti-NA sheep sera were determined by ELISA. Briefly, a 96-well microplate Immulon 2HB (Corning Inc, Lowell, USA) was coated with 50 μΙ/well of a 2.0 pg/ml VLP sample formulated in 25 mM tris-buffered saline (TBS), pH 7.2. The plate was incubated for 2 hr at room temperature (RT) with shaking and washed with 25 mM TBS buffer containing 0.05% Tween20 using a Microplate washer Biotek ELx405 (Biotek Inc, Winooski, USA). The plate was blocked with 150 pl/well of Superblock buffer (Pierce) for 30-40 min at RT. Six 2 or 4-fold serial serum dilutions starting from a 1 :10 dilution were prepared in Superblock buffer with 0.1 % Tween20. After blocking, 30 μΙ/well of each serum dilution was added to wells in duplicate. After 1 hr incubation, the plate was washed and filled with 30 μΙ/well of 50 ng/ml peroxide-conjugated donkey anti-sheep immunoglobulin (Jackson Immunoresearch Laboratories, West Grove, USA) diluted in Superblock buffer containing 0.1 % Tween20. After 1 hr incubation and washing, 100 μΙ/well of SureBlue Reserve TMB microwell peroxide substrate (KPL, Gaithersburg, USA) was added to each well. The plate was incubated for 10-15 min and the reaction was stopped by adding 25 μΙ/well of 1.0 M HCI. The plate was read at 450 nm on a microplate reader Infinity M200 (Tecan, Mannedorf, Switzerland). OD readings (OD450) were corrected by background subtraction and averaged.

[0135] Corrected mean OD450 values were plotted versus ln(serum dilution). Linear regression equation was generated using the linear portion of the serum titration curve. The endpoint Ab titer was determined by back calculating the serum dilution from In (serum dilution) which was obtained by extrapolating the linear regression line to cut-off absorbance OD450 = 0.2. The calculation was performed using the following equation:

[0136] Ab Titer = exp[(lnD2-lnD1 )^*(OD1 -0.2)/(OD1 -OD2) + lnD1]

[0137] where OD1 and OD2 represent OD values on linear portion of titration curve; D1 and D2 are the corresponding serum dilutions; 0.2 - cut-off absorbance value.

7. Neuraminidase inhibition assay

[0138] The principle of the NA inhibition assay includes testing NA activity in VLP or virus samples after their incubation with serial dilutions of serum containing NA-specific antibodies. The NA activity was measured by a modified fluorometric assay with 2'-(4- Methylumbelliferyl)-a-D-N-acetylneuraminic acid sodium salt hydrate (MUNANA) as a substrate and fluorometric detection of the fluorescent reaction product 4-Methyl Umbellipherone (MU) (Potier et al., 1979; Wetherall et al., 2003).

7.1 Reagent preparation

[0139] Stock solutions of 100 mM MUNANA (Gold Biotechnology Inc., St. Louis, USA) and 2.0 mM MU (Sigma-Aldrich, St. Louis, USA) were prepared in HyPure water (Fisher Scientific, Pittsburg, USA) in amber glass vials. The assay buffer 1 (AB1 ) composition included 32.5 mM MES sodium salt, 4.0 mM CaCI₂ and 0.1 % Tween 20 from Sigma with pH adjusted to 6.5. Assay buffer 2 (AB2) was prepared from AB1 by adding 60 pg/ml of bovine serum albumin (BSA) from Pierce. Stop solution was made using 0.1 M glycine and 25% ethanol from Fisher with pH adjusted to 10.7. MUNANA Stock solution was aliquot and stored at -20 °C. All other reagents were stored at 2-8 °C. Working solutions of 90 μΜ MU in AB1 and 300 μΜ MUNANA in AB2 were freshly prepared prior to performing the assay and used on the same day.

7.2 VLP sample preparation

[0140] VLP samples were diluted 1 :1 with 2X Stabilization buffer containing 60 mM MES- Na, 8.0 mM CaCL2, 1.0 M NaCL (Sigma), 0.1 % TPGS (Alpha Tocopheryl Polyethylene Glycol 1000 Succinate) (Cognis, Cincinnati, USA ) and 10% Glycerol (Fisher). The samples were aliquoted in 1.5 mL cryovials and stored at 2-8 °C for up to six months or at -20oC for up to one year. Prior to the test, VLP samples were diluted with AB1 to a final NA activity ranging from 1.0-2.0 nmol MU/well and used on the same day.

7.3 Assay procedure

[0141] The assay was performed in 96-well black microplates (Greiner Bio-One, Monroe, USA). MU standard solutions were prepared in AB1 and added to wells (90 μΙ/well) in duplicate to provide final MU concentrations of: 0, 0.1 , 0.5, 1 .0, 2.0 and 3.0 nmol/well.

[0142] Serum samples were stored frozen. Prior to the assay, the samples were thawed and stored at 2-8 °C until the assay was performed for up to one week. To provide 0 to 50% NA inhibition, two serum dilution schemes were used: six 4-fold serial dilutions within the range 1 :4 to 1 :4096 for elevated Ab titers and six 2-fold serial dilutions within the range 1 :2 to 1 :64 for low Ab titers. VLP sample (30 μΙ /well) was mixed with 30 μΙ of each serum dilution in duplicates or with 30 μΙ of AB1 in four replicates for VLP control. The plate was incubated at RT with shaking for 40 min. 30 μΙ of 300 μΜ MUNANA was added to wells containing VLP samples or 60 μΙ of AB1 (Substrate blank) to a final concentration of 100 μΜ. The plate was incubated at 37 °C with shaking on an incubator-shaker Jitterbug-4 (Boekel Scientific, Feasterville, USA) for 40 min. The reaction was stopped by adding 150 μΙ/well of Stop solution.

[0143] Fluorometric measurements were performed immediately with Modulus Microplate Multimode Reader (Turner Biosystems, Sunnyvale, USA) at Excitation 365 nm and Emission 410-460 nm and expressed in Relative Fluorescence Units (RFU).

7.4 Calculation of NA inhibition titer (NIT)

[0144] A standard curve RFU vs. MU and linear regression equation were generated using readings from standards corrected for blank fluorescence. Corrected sample RFU was obtained by subtracting the average substrate blank RFU from the measured mean sample RFU. The NA activity of the samples was calculated using the linear regression equation for MU standards and expressed as an amount of the product (nmol/well of MU) accumulated over incubation time. If needed, NA activity (NAA) was converted from nmol/well to mU/ml using the equation

[0145] NAA (mU/ml) = MU_x(nmol/well)*DF/(V_x*T_incub) = 0.56*MU x *DF

where MU is amount of MU accumulated over the enzymatic reaction; DF is dilution factor or final dilution of VLP sample; V_x = 30 pL volume of VLP sample; Tincub = 40 min incubation time; mU is milliUnit of enzyme activity corresponding to the conversion of 1 nmol substrate per 1 min.

[0146] The inhibition curve was generated as a % Residual NAA to the initial NAA for VLP control with no serum added versus log₂ [Serum Dilution] as shown in Figure 9. [0147] The optimized procedure included calculation of NAI titer corresponding to 25% of inhibition, or 75% of Residual NAA which was assigned NIT25 or just NIT. For its estimation, the linear part of the inhibition curve around 75% of Residual NAA was approximated by linear regression followed by the calculation of log₂ [Serum dilution] corresponding to 75% of Residual NAA. The NIT was reported as the serum dilution back calculated from antilog₂[NIT]. Any Log₂NIT < 0 was assigned a number 0 and corresponded to NIT = 1.

[0148] During assay development NAI titer was also estimated using two other cut-off values. The log₂NIT50 representing 50% of inhibition was calculated using the same inhibition curve. The log₂NIT10 representing approximately 10% of inhibition was defined based on (Sylte et al., 2007) as one log2 serum dilution below the log2 dilution corresponding to NAI <5%:

[0149] Log₂NIT10 = Log₂NIT5 - 1 .

8. Qualification of active NA in VLPs

[0150] Seven VLP samples with different subtypes of NA were prepared in stabilization buffer and tested for NAA, total protein and HA concentration and compared to other preparations of active NA including VLPs, purified enzymes, and inactivated virus vaccines. Stability of NAA in VLP was analyzed as % Residual NAA after storage at 2-8oC for 6 months.

9. Standardization of NA (in activity units) for Ab titration

[0151] Since neutralizing Ab titration requires a constant antigen concentration, it is important to establish an acceptable range for initial NA activity in VLP samples providing consistent neutralization Ab titer results. To set up this range, four dilutions of H3N2 A/Bris VLP sample with initial NAA = 3.21 nmol/well of MU were prepared as a neat sample, 2-fold, 3-fold and 4-fold dilutions. Each VLP sample was tested for NAI titer with H3N2-NA(A/Bris)- specific antiserum and a bias between log₂NIT values for neat and diluted samples was calculated.

10. Assay specificity

[0152] To compare NA inhibition in VLP sample by Ab specific to homologous and heterologous NA and HA, B-FI VLP was assayed for NIT using sheep serum samples specific to NA (B/FI), HA(B/FI), NA(H3N2 A/NY) and HA(H3N2 A/NY). To evaluate any interference effect of Ab to homologous HA and heterologous NA and HA on specific NA inhibition, different combinations of NA(B/FI)-specific antiserum with antisera specific to other NA and HA proteins were prepared and tested for NAI titer using a B-FI VLP sample. To analyze NA subtype specificity, antiserum specific to H5N1-NA (A/lndo) was tested for NIT against specific (H5N1 A/lndo) and nonspecific (H1 N1 A/NewCal, H3N2 A/NY and B/Sh) VLPs. [0153] Finally, to confirm NAI titer specific to only one subtype of NA in sera from subjects vaccinated by trivalent vaccine, 79 donors were tested for NIT before (preimmune) and after (postimmune) vaccination with a seasonal trivalent VLP vaccine containing H1 N1 A/Bris, H3N2 A/Bris, and B/FI influenza strains. Anti-NA immune response was tested for B and N2 subtypes of NA separately using B/FI and H3N2 A Br VLPs respectively and considered positive if post-to-pre Ratio of NAI titers (PPR) was > 2.0. The distribution of donors on groups with positive NAI response against both NA subtypes and against only one NA subtype was analyzed.

1 1. NAI assay validation

[0154] NAI assay validation was performed following ICH guidelines (ICH, 1996) and included testing assay accuracy, precision, and linearity.

1 1.1 Accuracy

[0155] Accuracy was measured using a dilution experiment in which serum from a ferret immunized with H3N2 A/Bris virus was serial diluted (2-fold) with preimmune (naive) serum up to a 1 :1024 dilution and tested for NAI titer using H3N2 A/Bris VLP. Expected NIT in diluted samples for both cut-off values NIT25 and NIT50 was calculated using the following equation:

[0156] NIT_dn = NIT_p/D + NIT_n*(D-1 )/D

[0157] where NIT_di|_, NIT_P and NIT_n are titers for diluted sample, positive (immune) and negative (preimmune) serums; D is positive serum dilution.

[0158] Expected log₂NIT10 for sample with dilution D was calculated as

[0159] log₂NIT10 (neat) - log₂D

[0160] To analyze any impact of different cut-off values on assay performance, accuracy was tested for all three levels of NA inhibition: 10% (log₂NIT10), 25% (log₂NIT25) and 50% (log₂NIT50) for bias between found and expected log₂ NIT values. Accuracy was considered acceptable if the absolute value for the calculated bias was < 1.0 which corresponds to one 2-fold serum dilution.

1 1.2 Intra- and inter-assay precision

[0161] The intra-assay precision was evaluated by testing three serum samples with different Ab titers corresponding to low, medium, and high levels of log₂NIT. These samples were prepared by diluting the anti H3N2-NA (A Bris) sheep serum with preimmune serum and tested six times each. To evaluate the inter-assay variability, NAI titer for ferret antiserum immunized with H3N2 A/Bris virus was tested by two analysts on two separate days using H3N2 A/Bris VLP. The intra-assay and inter-assay variation between Log₂NIT results, including between-day and between-analyst repeatability, was calculated by ANOVA and was presented as mean along with the %CV values. The % CV was acceptable if % CV< 10.0%.

1 1.3 Linearity and analytical range

[0162] Linearity was analyzed using data obtained for accuracy. The log₂NIT values measured from the assay were plotted against expected log₂ IT values and were subjected to linear regression analysis.

[0163] The linearity was acceptable if correlation coefficient value (R square) > 0.90. The range of the NAI assay was established as the range of log₂NIT values that met the acceptance criteria for accuracy, precision, and linearity.

12. Statistical analysis

All statistical analysis was performed by ANOVA using MS Office Excel software. I. Qualification of active NA in VLP samples

[0164] NA activity determined for seven VLP preparations with different subtypes of NA varied between 75 mU/ml and 480 mU/ml (Table 15). Relative NAA per mg of total protein was within the range of 0.08 to 1.23 mU/mg total protein and relative NAA per pg of HA was more consistent with only a 3.7-fold variability from 0.43 to 1 .61 mU/pg HA. All VLP preparations demonstrated high stability of active NA upon storage for 6 months at 2-8oC. %Residual NAA was within the range of 84-100%.

Table 15. Characterization of neuraminidase in VLPs

%Residual

Initial

VLP Lot NAA/TP NAA/HA NAA

Influenza virus strain NAA

# mU/mg mU/ug after 6 month mU/ml

storage at 2-8°C

B/Florida 04/2006 75508010 456 1.23 0.63 98.4

B/Brisbane 60/2008 714.22 88.0 0.14 1.18 100

A/Brisbane 10/2007 (H3N2) 7550801 1 426 1.18 0.55 91.1

A/Brisbane 10/2007 (H3N2) 081208 188 0.58 1.61 102

A/Brisbane 59/2007 (H1N1) XX10F001 480 1.20 0.73 96.2

A/California 04/2009 (H1N1) 09M001 250 0.08 1.44 87.6

A/Indonesia 05/2005 (H5N1) 0901 10 75.0 0.11 0.43 84

NA activity (NAA) was expressed as milliunits (nmol/min) per ml, mg of total protein (TP) and μg of hemagglutinin (HA).

Each number presented Mean value for 3 replicates with CV < 2.0%.

II. Standardization of NA (in activity units) for Ab titration [0165] Since measurable NAA in VLPs should be within the range for accuracy of MU measurement, the analytical range for fluorescence measurement of MU standards was validated first and found to be in the range of 0.1-3.0 nmol/well of MU. Four 2-fold serial dilutions of H3N2 A/Bris VLP with an initial NAA varying from 3.21 to 0.83 nmol/well were prepared and tested for NAI titer with H3N2-NA(A/Bris)-specific antiserum. NAI curves and the calculation of log₂NIT25 were shown in Fig.9. The log₂NIT tested for different initial NAA levels was between 7.83 to 8.48 with a bias which did not exceed 1.0 or one two-fold serum dilution (Table 16). Therefore, the initial NAA in VLPs was qualified within the range of 2.0 to 1.0 nmol /well to provide consistent testing of NIT.

Table 16. Consistency of NAI Titer tested at different VLP dilutions

Control NAA Bias:

VLP

(no serum) log2(N!T) Diluted- dilution

(NAA, nmol/well) Neat

neat 3.21 8.09 0

1 :2 1.74 8.33 0.24

1 :3 1.2 8.48 0.39

1 :4 0.83 7.83 -0.26

(VLP sample: A Brisbane/ 10/2007 (H3N2) Lot 7550801 1 ,

Antiserum sample: Ferret antiserum immunized with virus A/Brisbane/10/2007 (H3N2) virus)

III. Specificity for testing NAI in presence of homologous and heterologous NA and HA- specific antibody

[0166] The specificity of anti-NA and anti-HA sera was confirmed by testing total IgG titer by

ELISA. All four antisera demonstrated very high specificity for homologous proteins: B/FI-NA and B/FI-HA-specific sera had high titer against B/FI VLP and very low binding using H3N2

A/NY VLP. In contrast, H3N2 (A/NY)-NA and H3N2 (A/NY)-HA specific antisera had high binding with H3N2 A/NY VLP and low binding with B/FI VLP (Table 17).

Table 17. NA inhibition and Total IgG Titers tested against B/Florida VLPs with different combinations of sheep antiserums specific to NA (B/FL), HA (B/FL), NA(H3N2 A/NY) and HA(H3N2 A/NY) prepared by immunization with purified HA and NA proteins from B/Florida and H3N2 A/NY viruses

Antiserum Total IgG Total IgG

NA Inhibition Titer

Sample Titer Titer combination %Relative against against H3N2

NIT reactivity to B/Florida A/NY VLP adjusted inhibition by VLP (100 ng (100 ng log₂NIT

to 100 ng NA-specific protein/well) protein/well) protein/well antiserum

B-NA 13.23 34565 100.0 10368 1 1

NA inhibition by homoloj ^jous HA and heterologous NA and HA antibodies

B-HA 2.09 15.4 0.04 35900 1217

H3N2-NA 3.35 36.6 0.1 1 4485 14536

H3N2-HA 3.72 47.3 0.14 14 35916

H3N2-NA + H3N2-HA 4.66 91 .0 0.26 X X

B-HA + H3N2-NA +

X X

H3N2-HA 5.18 130.6 0.38

Impact of homologous HA and heterologous NA and HA antibodies on NA inhibition by homologous NA antibody

B-NA + B-HA 12.78 25320 73.3 X X

B-NA + H3N2-NA +

H3N2-HA 13.36 37892 109.6 X X

B-NA + B-HA + H3N2- NA + H3N2-HA 13.26 35196 101.8 X X

x - not applicable

[0167] Confirming the specificity of the assay, a high NAI titer was detected when testing B/FI VLP with homologous B-NA antiserum, while NITs for homologous B-HA and heterologous H3N2-NA and H3N2-HA antisera were 735-2300 times lower. A mixture of H3N2-NA and H3N2-HA antisera or a mixture of B-HA, H3N2-NA and H3N2-HA antisera resulted in a low total NIT with B/FI VLP matching the sum of the NAI titers from individual serum samples.

[0168] To test any interference effect, B-NA antiserum was combined with different combinations of homologous B-HA antiserum and heterologous H3N2-NA and H3N2-HA which simulates antiserum from immunization with B/FI VLP alone or with a combination of B/FI and H3N2 VLPs. The log₂NIT values obtained for these simulated antisera with additional B-HA, H3N2-NA and H3N2-HA antibodies had no significant difference from log₂NIT tested for B-NA antiserum alone: the calculated bias between log₂NIT titers for original B-NA antiserum and any of its modification did not exceed 0.5. It confirms that specific NAI antibody is detected by VLP NA in the presence of heterologous NA antibody and VLP NA is not neutralized by HA antibody.

[0169] Specificity of NAI assay was also confirmed by testing NIT against one subtype of NA (H5N1 A/lndo) using VLPs with different NA subtypes. This antiserum specifically inhibited the homologous subtype of NA in H5N1 A/lndo VLP and did not inhibit other subtypes such as H3N2 A/NY and B/FI (Fig 10). For another strain of N1 subtype NA (H1 N1 A/NewCal) a reduced inhibition titer was detected. Cross-reactivity of anti-NA (H5N1 A/lndo) antibodies towards different subtypes of NA was evaluated by the ratio of the NAI titer against a heterologous VLP to the NIT against the homologous H5N1 A/lndo VLP (Table 18). The cross-reactivity of H1 N1 -NA (A/NewCal) to H5N1 -NA specific Ab was low but detectable

(2.16%), while cross-reactivity of N2 and B subtypes of NA to the same Ab was not detected.

Table 18. Cross-reactivity of H5N1 NA antibodies with different types of

Neuraminidase tested with NA inhibition assay

VLP NIT %Cross-reactivity

H5N1 A/Indonesia Lot 022007 8462.7 N/A

H 1N1 A/NewCaledonia Lot 75508005 182.7 2.16

H3N2 A/NewYork Lot 090407 1.0 0.01

B/Shanghai Lot 101607 4.6 0.05

[0170] NAI titers measured independently against B and N2 subtypes of NA in sera from donors immunized with the trivalent influenza VLP vaccine demonstrated no correlation (R = 0.06) between immune responses for these two NA subtypes. While 66% of donors had a positive immune response (PPR≥2.0) to both NA subtypes, 24% of donors had a positive immune response against only one subtype of NA with 19% against B-NA and 5% against N2-NA (Table 5). This demonstrates that the assay is able to identify immune response to individual NA subtypes when testing human vaccines for anti-NA Ab responses.

IV. Accuracy

[0171] Reducing a NA-specific Ab concentration in serum samples by dilution of immune serum with preimmune (na^'ive) serum led to a substantial decrease of the inhibition effect

(Fig. 1 1 ). Choosing different points on the inhibition curve for undiluted immunized serum resulted in a different value for NAI titer which varied from a log₂NIT10 of 1 1.0 (at 10% of NA inhibition) to log₂NIT50 of 5.65 (at 50% of NA inhibition), (Table 19).

Table 19. Distribution of donors vaccinated with 60 μg HA dose of seasonal trivalent VLP vaccine (A/Brisbane/59/2007(HlNl), A/Brisbane/ 10/2007 (H3N2),

B/Florida/04/2006) on groups with no (Post-to-Pre Ratio of NAI titers < 2.0) and positive (PPR > 2.0) NAI immune response against both B-NA and H3N2-NA and only one B-NA or H3N2-NA

Geometric Mean for Post- to-Pre Ratio (PPR) of NAI

%Total

Group N titers

N

B-NA H3N2-NA

Total 79 100 9.4 5.9 Positive response for both B-NA and H3N2-NA 52 65.8 15.2 1 1.6

Positive response only for B-NA 15 19 1 1.7 1.4

Positive response only for H3N2-NA 4 5.1 1.4 5.0

No response for both B-NA and H3N2-NA 8 10.1 1.2 1.1

Table 20. Accuracy of log₂NIT10, log₂NIT25 and log₂NIT50 tested in serial dilutions of immune anti H3N2 A/Brisbane ferret serum diluted with preimmune serum.

(For preimmune serum log₂NIT25 = 0.87 and log₂NIT50 = 0)

Antiserum log₂ NIT10 2 NIT25 12 NIT50 Dilution

Expected Found Bias Expected Found Bias Expected Found Bias

Neat 1 1 .0 1 1.0 0.0 8.58 8.58 0.00 5.65 5.65 0.00

2 10.0 10.0 0.0 7.59 8.24 0.65 4.65 4.03 -0.62

4 9.0 10.0 1 .0 6.60 6.23 -0.37 3.65 2.07 -1.59

8 8.0 9.0 1.0 5.63 4.84 -0.79 2.65 0.98 -1.67

16 7.0 7.0 0.0 4.68 3.65 -1.03 1.65 0.07 -1.59

32 6.0 9.0 3.0 3.78 3.32 -0.46 0.65 -2.29 -2.94

64 5.0 7.0 2.0 2.96 1.97 -0.99

128 4.0 5.0 1.0 2.26 1.24 -1.02

256 3.0 5.0 2.0 1 .73 2.71 0.98

512 2.0 3.0 1.0 1 .36 0.88 -0.48

1024 1.0 3.0 2.0 1 .14 0.0 -1.14

Bias > 1.0 was shown in bold numbers

[0172] For three NAI cut-off levels (10%, 25% and 50%), the accuracy range was determined as a range of serum dilutions corresponding to diluted samples with NAI titers with bias≤1.0 between measured and expected log₂NIT. Log₂NIT50 provided a very narrow range with acceptable accuracy due to a bias≥ 1.6 at 1 :8 and higher serum dilutions. A broader accuracy range, up to 1 :16 serum dilution, was found for using log₂NIT10 or 10% cut-off level. However, 25% cut-off level and calculation of Log₂NIT25 tremendously expanded this range up to 1 :512 serum dilution. Based on the best accuracy range, log₂NIT25 designated also as log₂NIT was chosen for final NAI titer assignment and further assay validation. V. Intra-assay and inter-assay precision

[0173] Intra-assay repeatability was analyzed at three levels of log₂NIT25, ranging from 7.86 to 3.41 and demonstrated a low variability, with % CV≤ 2.3% (Table 21 ). Inter-assay precision measured as total variability of mean log₂NIT on two different days performed by two analysts corresponded to % CV of 3.95%. Data collected for inter-assay precision were arranged into groups representing between-day and between-analyst variability (Table 22).

Between-day variability demonstrated % CV of 2.55% and a statistically insignificant bias of

0.31 (p-value > 0.05 for Student t-test). Between-analyst variability had a %CV of 3.78% and bias of 0.46 (p-value < 0.05). Overall, NAI assay demonstrated excellent precision since %

CV and bias were well below the acceptance criteria for % CV (≤ 10%) and bias (≤ 1.0).

Table 21. Intra-assay precision for testing Log₂NIT with H3N2 A/Brisbane VLPs and anti H3N2-NA (A/Brisbane) sheep serum

Diluted 1 :8 with preimmune serum middle 5.37 2.3

Diluted 1 : 128 with preimmune serum low 3.41 1.3

Table 22. Between-day and between-analyst assay repeatability for testing log₂NIT with anti H3N2-NA (A/Brisbane) ferret serum.

p-value

Analyst/Day # 1 2 Mean Bias for t-test

Between-Day

Mean (n=6) 8.76 8.45 8.61 0.31 0.13

%CV 2.91 4.73 2.55

Between- Analyst

Mean (n=6) 8.84 8.38 8.61 0.46 0.02

%CV 2.41 3.98 3.78

VI. Linearity and analytical range

[0174] Using data from the accuracy testing, the measured log₂NIT was plotted against expected log₂NIT and linear regression analysis was performed (Fig. 12). The NAI assay produced a good linear response over the entire range of log₂NIT from 8.6 to 1 .4 and also demonstrated a high correlation between both parameters with R square of 0.94. Overall, the regression line fit well to the line of identity with variation only at the lower end of the range, where lower than expected values for measured log₂NIT were obtained. Therefore, the analytical range for NAI assay was determined to be 1.4 to 8.6 of log₂NIT calculated for

25% of NA inhibition which met the acceptance criteria for accuracy, precision, and linearity.

[0175] The analytical performance of the NAI assay is summarized in Table 23. Since all acceptance criteria were met, the assay was validated and can now be used for long-term clinical studies to measure NAI antibody responses to influenza vaccines.

Table 23. Analytical Performance characteristics for testing NAI Titer as log₂NIT25 using VLP samples

Parameter Acceptance Criteria Results

NA subtype %Cross-

%Cross-reactivity = 0

reactivity < 10%

Specificity Interference with anti- HA AB in VLPs < %Relative reactivity < 4.2% 10.0%

For log₂NIT = 7.9 %CV = 2.0

Intra-assay Repeatability For log₂NIT = 5.4 %CV =

%CV < 10.0

(Precision) 2.3

For log₂NIT = 3.4 %CV = 1.3

Inter-assay Reproducibility

(Precision) %CV < 10.0 %CV < 4.0

%CV < 10%; Bias <

Between-Day Reproducibility 1 .0 %CV < 3.0% Bias < 0.4

%CV < 10%; Bias <

Between-Analyst Reproducibility 1.0 %CV < 4.0% Bias < 0.5

Accuracy Bias < 1.0 Bias < 1.0

Linearity Rsquare > 0.90 Rsquare = 0.94

Analytical Range for log₂NIT 1.36-8.58

Analytical Range for NIT 2.6-382

Discussion

[0176] It is widely recognized that generating protective immune responses against both HA and NA viral proteins significantly improves efficacy of influenza vaccine (Sylte and Suarez, 2010; Bright at al., 2008, 2007). Therefore, there is a significant demand for a validated assay to measure NA neutralizing antibody response after vaccination. Key parameters for such an assay must include 1 ) a stable source of active NA enzyme; 2) standardization of an active NA sample used for Ab titration; and 3) high reactivity to specific NA antibodies with little to no reactivity against HA antibody and heterologous NA antibodies. To establish the analytical range, assay accuracy, precision, and linearity must be validated.

[0177] The results for NAA in VLPs of the present invention were the first time it's activity was quantified in enzymatic units and allowed comparison with NAA in inactivated vaccines and purified NA preparations. NAA in VLPs herein varied from 0.43 to 1.61 rnll/pg HA and corresponded to the highest NA activity found in inactivated influenza vaccines (1.0-1.6 mU/pg HA) (Lambre et al., 1989) while most of these vaccines had much lower NA activity (Kendal et al., 1980; Chaloupka et al., 1996). NAA in VLPs exceeded enzymatic activity reported for commercial purified H1 N1 NA (167 mU/mL) and H5N1 NA (60 mll/mL) from SinoBiological (Beijing, China). Assuming that the NA content in VLPs is less than 10% of the total protein, H5N1 VLPs had about 2-fold lower NA activity per NA protein than purified H5N1 NA (2.5 mU/pg) from RnD Systems (Minneapolis, USA). Overall, NA activity in VLPs can potentially match the highest level of NA activity detected for inactivated vaccines and purified enzyme preparations.

[0178] The inventors demonstrated that six strains of VLP prepared in stabilization buffer and stored at 2-8°C for 6 months had maintained 84-100% of NA activity. The active NA was more stable in VLPs than in purified NA preparations and inactivated vaccines. By vendor specifications, all commercial purified NA preparations were stable up to 6-12 months only upon storage at -20°C. Based on limited published results, H1 N1 inactivated vaccines showed a loss of 50-100% of NA activity after 6 months of refrigerated storage (Kendal et al., 1980). The VLP technology herein therefore provided different strains of VLPs with the most active and stable NA available which can be considered the best source of active enzyme for NA neutralizing Ab assay. Another potential advantage of VLP could be maintenance of the native NA structure allowing better binding of Ab. From preliminary experiments, anti-NA Ab from serum samples generated by immunization with VLP vaccine had higher affinity to homologous NA presented in VLPs than to purified soluble NA (data are not presented).

[0179] All known NA neutralizing Ab studies had used a non-qualified source of active NA evaluated only by its relative activity, and none of published NAI Ab titers can be compared. In the present study, the inventors had made the first attempt at setting up an NA standard for Ab titration. Since preparing VLP samples with equal NA activity is problematic, the range for the acceptable NA activity was established between 1.0-2.0 nmol/well of ML). In this range the assay produced similar results for NAI titers with a calculated bias less than one 2- fold serum dilution. Using an average value for standard NA activity (1.5 nmol/well), each Ab titer can be defined as a serum dilution providing a specified level of inhibition for 25 pU of active NA. Various VLP preparations for similar and different influenza strains could have a wide range of NA activity, which could lead to a variable number of NA molecules used for Ab titration. For given sample, that lost 80% of NA activity during storage, normalization by the number of activity units provides 5 times higher number of NA molecules and would generate a much lower Ab titer than for testing using a 100% active NA. This demonstrates the crucial importance of NA stability for getting reproducible NIT results. To ensure the same NA activity in VLP sample for testing pre- and postimmune serum samples, the inventors ran them on one microplate. To compare Ab titers specific to different subtypes of NA, additional NA standardization in VLPs by both enzymatic activity and concentration needs to be developed. From this standpoint, evaluation of NA immune response using a seroconversion ratio of Ab titers for post- and preimmune sera seems to be a more reliable approach than measuring Ab titers themselves.

[0180] Generally, binding HA-specific Ab with HA can partially inhibit access of substrate or anti-NA Ab to NA due to the proximity of both surface proteins, which could result in a misleading NAl titer. The inhibition by anti-HA Ab binding can be avoided by using reassortant viruses with mismatched HA which, for example, has been used for the NAl assay using the glycoprotein fetuin as a substrate (Hassantoufighi et al., 2010, Cate et al., 2010). The big advantage of a synthetic substrate such as MUNANA with a low molecular weight, is that the NAl assay has shown no interference from homologous HA-specific Ab. Furthermore, heterologous HA and NA-specific Ab also had no significant impact on NAl titer. The assay demonstrated excellent specificity to NA subtypes. The NAl titers for serums from donors immunized with the trivalent VLP vaccine confirmed assay specificity and led to the significant conclusions that patients immunized with a polyvalent vaccine generated an independent immune response against each subtype of NA and the NAl assay is able to detect this independent and specific immune response. It should be noted that the high specificity of MUNANA-based assay may result in its ability to detect only part of NA-specific antibodies which directly bind with the active site of the enzyme.

[0181] The cut-off value on NAl inhibition curve used for Ab titer estimation appeared to be a key factor in determining the accuracy range and the overall analytical range of the assay. The evaluation of NAl titer using 25% of NA inhibition level provided a significantly wider accuracy range than for previously used 10% and 50% cut-off levels.

[0182] Both intra-assay precision evaluated at low, medium, and high levels of log₂NIT25 and inter-assay precision examined by between-day and between-analyst variability demonstrated excellent assay performance with all %CV values below 5.0%. The high assay precision was a result of the combination of high NA stability, standardization of NA (in activity units) used for Ab titration and optimization of the cut-off value for Ab titer estimation. The high assay reproducibility demonstrates its suitability for conducting long-term multisite clinical trials.

[0183] Finally, regression analysis of measured versus expected log₂NIT values confirmed linearity of the assay with an R square value of 0.94. The assay has been validated for testing NA neutralizing Ab in serum samples within the range 1.4 to 8.6 of log₂NIT. The upper limit of the range could be greater if antiserum with a higher Ab titer was evaluated. For example, Ab Titer with log₂NIT up to 12.8 was detected in some clinical samples from donors immunized with trivalent VLP vaccine. To knowledge of the inventors, this study is the first one presenting the validation of a NA neutralizing Ab assay. With this data, the reported analytical range could be used as a benchmark for all currently known assays and any new assay development.

[0184] The new NA neutralizing Ab assay using VLPs as a substitute for live virus could be of great value in further studies of NA immunity, especially with regards to new research over synergistic effect of NA and HA-specific Ab's on protection against influenza virus infection (Bosch et al., 2010).

[0185] The foregoing detailed description has been given for clearness of understanding only and no unnecessary limitations should be understood there from as modifications will be obvious to those skilled in the art.

[0186] While the invention has been described in connection with specific embodiments thereof, it will be understood that it is capable of further modifications and this application is intended to cover any variations, uses, or adaptations of the invention following, in general, the principles of the invention and including such departures from the present disclosure as come within known or customary practice within the art to which the invention pertains and as may be applied to the essential features hereinbefore set forth and as follows in the scope of the appended claims.

[0187] The disclosures, including the claims, figures and/or drawings, of each and every patent, patent application, and publication cited herein are hereby incorporated herein by reference in their entireties.

References

Bosch B.J., Bodewes R., de Vries R.P., Kreijtz J.H., Bartelink W., van Amerongen G., Rimmelzwaan G.F., de Haan C.A., Osterhaus A.D., Rottier P.J. 2010. Recombinant soluble, multimeric HA and NA exhibit distinctive types of protection against pandemic swine-origin 2009 A(H1 N1 ) influenza virus infection in ferrets. J.Virol. 84, 10366-10374.

Bright R.A., Carter D.M., Crevar C.J., Toapanta F.R, Steckbeck J.D., Cole K.S., Kumar N.J., Pushko P., Smith G., Tumpey T.M., Ross T.M. 2008. Cross-clade Protective Immune Responses to Influenza Viruses with H5N1 HA and NA elicited by an Influenza Virus-Like Particle. PLoS ONE, 1 , e1501.

Bright R.A., Carter D.M., Daniluk S, Toapanta F.R, Ahmad A, Gavrilov V, Massare M,

Pushko P., Mytle N., Rowe T., Smith G., Ross T.M. 2007. Influenza virus-like particles elicit broader immune responses than whole virion inactivated influenza virus or recombinant hemagglutinin. Vaccine 25, 3871-3878.

Cate T.R., Rayford Y., Nino D.,Winokur P., Brady R., Belshe R., Chen W., Atmar R.L., Couch R.B. 2010. A high dosage influenza vaccine induced significantly more neuraminidase antibody than standard vaccine among elderly subjects. Vaccine 28, 2076- 2079.

Chaloupka I., Schuler A., Marschall M., Meier-Ewert H. 1996. Eur. J Clin Microbiol Infect Dis, 15, 121-127.

Feng, J., Gulati, U., Zhang, X.,Keitel, W.A., Thompson, D.M., James, J.A., Thompson, L.F., Air, G.M. 2009. Antibody quantity versus quality after influenza vaccination. Vaccine 27, 6358-6362.

Gubareva LV., Webster RG., Hayden FG. 2002. Detection of influenza virus resistance to neuraminidase inhibitors by an enzyme inhibition assay. Antivir.Res. 53, 47-61 .

Hassantoufighi A., Zhang H., Sandbulte M., Gao J., Manishewitz J., King L, Golding H., Straight T.M., Eichelberger M.C. 2010. A practical influenza neutralization assay to simultaneously quantify hemagglutinin and neuraminidase-inhibiting antibody responses. Vaccine, 28, 790-797.

International Conference on Harmonization (ICH) of Technical Requirements for the Registration of Pharmaceuticals for Human Use, Validation of analytical procedures: Methodology, adopted in 1996, Geneva

Johansson B.E., Bucher D.J., Kilbourne E.D. 1989. Purified influenza virus hemagglutinin and neuraminidase are equivalent in stimulation of antibody response but induce contrasting types of immunity to infection. J.Virol. 63, 1239-1246.

Kalbfuss B., Knochlein A., Krober T., Reichl U. 2008. Monitoring influenza virus content in vaccine production: precise assays for the quantitation of hemagglutination and neuraminidase activity. Biologicals, 36, 145-161.

Kang S., Song J., Quan F., Compans R.W. 2009. Influenza vaccines based on virus-like particles. Vir.Res. 143, 140-146.

Kendal A.P., Bozeman F.M., Ennis F.A. 1980. Further studies of the neuraminidase content of inactivated influenza vaccines and the neuraminidase antibody responses after vaccination of immunologically primed and unprimed populations. Infection & Immunity, 29, 966-971. Lai J.C.C., Chan W.W.L., Kien F.,NicholIs J.M., Malik Peiris J.S., Garcia J.M. 2010. Formation of virus-like particles from human cell lines exclusively expressing influenza neuraminidase. J.Gen.Vir. 91 , 2322-2330.

Lambre C.R., Chauvaux S., Pilatte Y. 1989. Fluorometric assay for the measurement of viral neuraminidase in influenza vaccines. Vaccine, 7, 104-105.

Mahmood K., Bright RA, Mytle N., Carter DM., Crevar CJ., Achenbach JE., Heaton PM., Tumpey TM., Ross TM. 2008. H5N1 VLP vaccine induced protection in ferrets against lethal challenge with highly pathogenic H5N1 influenza viruses. Vaccine 26, 5393-5399.

Potier M., Mameli L, Belisle M., Dallaire L, Melancon SB. 1979. Flurometric Assay of Neuraminidase with a Sodium (4-methylumbelliferyl-a-D-N-Acetylneuraminate) substrate. Anal.Biochem. 94, 287-296.

Pushko P, Tumpey TM, Bu F, Knell J, Robinson R, Smith G. 2005. Influenza virus-like particles comprised of the HA, NA, and M1 proteins of H9N2 influenza virus

induce protective immune responses in BALB/c mice. Vaccine 23, 5751-5759.

Rowe T., Abernathy R.A., Hu-Primmer J., Thompson W.W., Lu X., Lim W., Fukuda K., Cox N.J., Katz J.M. 1999. Detection of Antibody to avian Influenza A (H5N1 ) Virus in Human Serum by Using a Combination of serologic Assays.

J Clin Microbiology 37, 937-943

Sylte,M.J.,Suarez, D.L., 2010. Influenza neuraminidase as a vaccine antigen, in: Compans, R.W.,Orenstein, W.A. (Eds.), Vaccines for Pandemic Influenza, Current Topics in Microbiology and Immunology 333., Springer-Verlag, Berlin Heidelberg, pp.227-241.

Sylte.M.J., Hubby, B.,Suarez, D.L. 2007. Influenza neuraminidase antibodies provide partial protection for chickens against high pathogenic avian influenza infection. Vaccine 25, 3763- 3772.

Wetherall NT, Trivedi T., Zeller J.,Hodges-Savola C, McKimm-Breshkin JL, Zambon M., Hayden FG. 2003. Evaluation of neuraminidase enzyme assays using different substrates to measure susceptibility of influenza virus clinical isolates to neuraminidase inhibitors: Report of the Neuraminidase Inhibitor Susceptibility Network. J. Clin. Microbiol. 41 , 742-750.

Claims

CLAIMS:

1. A method of determining influenza antibody activity in a sample, comprising:

incubating the sample with an influenza VLP comprising at least one hemagglutinin (HA) or neuraminidase (NA) moiety; and

measuring influenza HA or NA activity in the sample.

2. The method of claim 1 , wherein antibodies in the sample neutralize HA or NA activity in the VLP.

3. The method of claim 1 , wherein antibodies in the sample binds to the HA or NA.

4. The method of claim 1 , wherein the HA or NA is from a single subtype of influenza virus.

5. The method of claim 1 , wherein the VLP comprises HA or NA from multiple subtypes.

6. The method of claim 1 , wherein the NA is from an influenza A, influenza B or influenza C virus.

7. The method of claim 6, wherein the influenza virus is H1 N1.

8. The method of claim 6, wherein the NA subtype is selected from the group consisting of N1-N9.

9. The method of claim 1 , wherein the HA is from an influenza A, influenza B or influenza C virus.

10. The method of claim 9, wherein the influenza virus is H1 N1.

1 1. The method of claim 9, wherein the HA subtype is selected from the group consisting of H1 -H16.

12. The method of claim 9, wherein measuring the HA activity comprises incubating the sample with red blood cells.

13. The method of claim 1 , wherein measuring the NA activity comprises incubating the sample with an NA substrate.

14. The method of claim 13, wherein exposure of the substrate to NA results in release of a detectable reaction product.

15. The method of claim 14, wherein the reaction product is fluorescent.

16. The method of claim 15, wherein the NA substrate is 2'-(4-Methylumbelliferyl)-a-D-N- acetylneuraminic acid sodium salt hydrate (MUNANA) or a derivative thereof.

17. The method of claim 16, wherein the reaction product is 4-Methyl Umbellipherone (MU).

18. The method of claim 14, further comprising calculating a NA inhibition titer.

19. The method of claim 1 , wherein more than one VLP type is incubated with the sample.

20. The method of claim 19, wherein the sample is incubated with a VLP in at least one well.

21. The method of claim 20, wherein one VLP type is incubated with the sample in each well.

22. The method of claim 21 , wherein each VLP comprises a different NA subtype.

23. The method of claim 1 , wherein the sample is a serum sample.

24. The method of claim 23, wherein the serum sample is a human serum sample.

25. The method of claim 24, wherein the human has been vaccinated.

26. The method of claim 25, wherein the vaccine is an inactivated influenza vaccine.

27. The method of claim 25, wherein the vaccine is an influenza VLP vaccine.

28. The method of claim 25, wherein the vaccine is a live influenza vaccine.

29. The method of claim , wherein the VLP retains at least 80% NA activity when stored for 6 months.

30. The method of claim 1 , wherein the VLP comprises an influenza M1 protein comprising a YKKL sequence (SEQ ID NO: 1 ) in the late domain.