RU2186111C2 - Recombinant neuraminidase of influenza type na2 (variants) - Google Patents

Abstract

FIELD: genetic engineering, biochemistry, enzymes. SUBSTANCE: recombinant neuraminidase of influenza type NA2 is prepared by culturing insect infected cells or yeast cells in a suitable medium. Insect cells are infected with virus transformed by homologous recombination of its genome using the expression vector pPP1IVNAflS. Recombinant neuraminidase is used in vaccine against influenza. Invention provides to prepare vaccine showing effectiveness against influenza virus type NA2. EFFECT: improved method of preparing, enhanced effectiveness of vaccine. 3 cl, 20 dwg, 2 tbl, 3 ex

Description

 The invention relates to a recombinant influenza neuraminidase, an expression vector by which recombinant neuraminidase can be expressed in host cells by the production and purification of recombinant neuraminidase, influenza vaccines and the use of recombinant neuraminidase in accordance with the present invention.

 Influenza epidemics caused by viruses A and B lead to significant discomfort for the people affected by them and have a very big impact on social and economic life. Influenza causes a high mortality rate in the elderly and patients suffering from chronic diseases. Since the use of inactivated vaccines based on viral material grown in chicken embryos in the 40s of our century, they have shown their obvious effectiveness against influenza and have led to a significant reduction in mortality among people at risk.

 Influenza viruses are unique among other viruses developing in the bronchioles, as they undergo significant antigenic changes (the so-called “drift”) in their two surface antigens, namely hemagglutinin (HA) and neuraminidase (NA).

 In addition, influenza A due to the phenomenon of "shift" (shift) is able to bypass common immunity. The virus that occurs in humans is a neuraminidase gene that appears from animal sources of influenza genes. In 1957, the most widespread virus of the type NA1 at that time was replaced by a new virus of the type NA2. Since 1977, type NA1 viruses have returned to the human population. Current vaccines should therefore preferably be directed against both types of NA1 and NA2 viruses.

 Neuraminidase catalyzes the removal of terminal sialic acid residues of glycosyl groups, while the potential receptors for hemagglutinin are destroyed (Gottschalk 1957; Burnet and Stone, 1947). It is believed that neuraminidase plays an important role in inhibiting virus aggregation and inhibiting its ability to spread from cell to cell (Colman and Ward, 1985).

Each neuraminidase molecule (M _r = 240000) has a structure that resembles a toadstool and consists of four identical polypeptide chains formed by two dimers that are linked by disulfide bridges and are further bonded to each other by non-covalent bonds (Bucher and Kilbourne, 1972; Laver and Valentine, 1969; Vargese et al., 1983; Ward et al., 1983). In contrast to hemagglutinin, neuraminidase is fixed in the lipid membrane using the non-spliced NA-terminal lipophilic sequence (Fields et al., 1983; Blosk et al., 1982), the so-called membrane anchor. The largest part of the entire structure protrudes beyond the membrane and forms there a peripheral “head” part, which has the shape of a box, which is placed over an elongated region of the “stem” (Wrigley et al., 1973). The head contains a monomer that has its own catalytic site and contains at least four NA-linked glycosyl groups (Colman et al., 1983; Ward et al., 1982). The presence of O-glycosylation has been established recently.

 Due to their external location, hemagglutinin and neuraminidase antigens are the most important targets in the structure of the virus for the host immune system. As for antigens that specifically bind to hemagglutinin, it is believed that they neutralize the invasive ability of the virus, probably by blocking the early stages of infection (Hirst, 1942; Kida et al., 1983). Neuraminidase-specific antibodies usually do not prevent primary infection of the target cell (Jahiel and Kilbourne, 1966; Kilbourne et al., 1968; Johanssen et al., 1988), but they block the spread of the virus. Further, due to the occurrence of competitive processes, the immune response to neuraminidase is apparently partially suppressed in favor of the more common hemagglutinin antigen (Johanssen et al., 1987; Kilbourne, 1976). As a result, the effects of neuraminidase immunity are usually overshadowed by antibodies that neutralize hemagglutinin. For this reason, the attention of vaccine developers for a long time has been almost exclusively focused on hemagglutinin.

 However, a number of experimental observations indicate that neuraminidase is actually able to play an important role in creating protective immunity against influenza (Schulman et al., 1968; Johansen and Kilbourne, 1990; Johansen et al., 1993). Basic research on the immunogenic potential of neuraminidase requires very pure antigens in sufficient quantities and with the correct three-dimensional structure. To date, neuraminidase is obtained by treating the envelopes of the virus with detergents (Gallgher et al., 1984; Kilbourne et al., 1968) or by proteolytic cleavage of the protein head, often using pronase (Seto et al., 1966; Rott et al., 1974), followed by purification of neuraminidase. Although these methods are applicable to some extent, they have significant limitations in terms of yield and purity of the product.

 In this regard, the aim of the present invention is a recombinant influenza neuraminidase, which has antigenic properties corresponding to the properties of natural neuraminidase and is correctly folded.

In accordance with the present invention, the indicated recombinant neuraminidase in substantially purified form can be obtained:
a) culturing in a culture medium host cells transformed with a neuraminidase expression vector or infected with a virus that is transformed with a neuraminidase expression vector, the expression vector comprising at least a portion of the coding region of the influenza virus neuraminidase gene minus the region encoding the membrane anchor, or a modified version thereof, which is preceded by a signal sequence connected in phase; and
b) isolation of neuraminidase expression products from the culture medium.

 The recombinant neuraminidase of the present invention, secreted into the culture medium, can, for example, be used in basic studies in which a separate vaccination with neuraminidase is carried out in order to determine the role of neuraminidase in the vaccine. In practice, recombinant neuraminidase will probably still be used in combination with hemagglutinin to increase the degree of protection (percentage of the vaccinated population that is effectively protected against infection) and resistance to protection (protection against later epidermal strains).

 In the present invention, in particular, a recombinant NA2 influenza neuraminidase is claimed, which can be obtained by culturing host cells in a suitable growth medium and isolating the neuraminidase expression product from the growth medium. In practice, this procedure involves transferring the recombinant expression module from pAC21VNAS to wild-type baculovirus or its derivative. The host cells are then infected with this recombinant baculovirus.

 Host cells that are used to produce recombinant influenza neuraminidase are predominantly derived from lower eukaryotic organisms such as insects, mainly from the insect sf9 cell line, but yeast cells such as Saccharomyces or Pichia can also be used.

 The present invention also relates to two vectors for the expression of a secreted influenza neuraminidase containing a replication initiation site, at least a portion of the coding region of the influenza neuraminidase gene minus the region that encodes the membrane anchor, or a modified version thereof, a signal sequence located at the 5'-position from the coding region and associated in phase, a promoter located at the 5'-position from the signal sequence, and a transcription terminator located at the 3'-position of the coding region. The invention, in particular, claims a vector for the expression of the NA2 influenza-secreting neuraminidase that contains a replication initiation site encoding a region of a gene of the A / Victoria / 3/75 influenza strain minus a region encoding a membrane anchor, or a modified variant thereof.

 For expression in insect cells, this vector is placed in a cell together with wild-type baculovirus or its derivative. Due to double homologous recombination, a recombinant baculovirus is obtained, wherein the expression module from the vector is introduced into the genome of the virus. After plaque purification, a stock of recombinant baculoviruses is obtained, which can then be used, for example, to infect sf9 cells.

 Preferably, the signal sequence of the hemagglutinin gene of the A / Victoria / 3/75 (H3N2) influenza virus NA2 is used. The invention mainly includes the vector pAC21VNAS, registered January 3, 1994 in the Plasmid Collection of the Laboratory of Molecular Biology (Laboratorium voor Moleculaire Biologie - Plasmidencollectie (LMBP)) at K.L. Ladeganckstraat 35, B-9000 Ghent, Belgium, and received the deposit registration number LMBP 2976, which is used to transform a virus, such as, for example, baculovirus, by double homologous recombination. In accordance with the present invention, a vector expression module including transcriptional regulation signals, a signal sequence and a coding region is inserted into the genome of the virus.

 In yet another embodiment of the present invention, a second vector of the present invention is used. This vector is designed for use in yeast and includes, for example, a replication initiation site encoding the region of the neuraminidase gene of strain A / Victoria / 3/75 of influenza NA2 minus the regions that encode the membrane anchor and stem part of neuraminidase, respectively, or its modified variants, the signal sequence located at the 5'-position from the coding region and associated with it in phase, a promoter located at the 5'-position from the signal sequence, and a transcription terminator located at the 3'-position from the coding uyuschey area.

 The promoter and terminator sequences are preferably homologous and are isolated from Pichia pastoris methylotrophic yeasts such as the alcohol oxidase 1 gene sequences. For example, the signal sequence contains a signal for the secretion of the prepro factor, α Saccharomyces cerevisiae mating.

 The indicated vector pPP1IVNAf1s was registered on January 3, 1995 in Laboratorium voor Moleculaire Biologie - Plasmidencollectie (LMBP), K.L. Ledeganckstraat 35, B-9000 Ghent, Belgium, and received an LMBP 3223 registration number upon deposit.

 The recombinant neuraminidase of the present invention has proven capable of generating protective immunity against influenza viruses, in particular influenza viruses of the type NA2. The invention also relates to a flu vaccine that contains recombinant neuraminidase.

 In addition, the invention relates to a method for producing recombinant neuraminidase and to a method for its purification.

 In the description of the present invention and in the formula, the term "NAS" refers to secreted (recombinant) neuraminidase; "pNA" refers to natural pronase-treated neuraminidase; "NA" means neuraminidase.

 The present invention is further illustrated by the following examples, which are provided only to illustrate and not to limit the scope of the claims of the present invention.

Example 1. Expression, purification and study of the properties of recombinant influenza neuraminidase NA2
Materials and methods
1. Construction of a gene encoding secreted neuraminidase, and its integration into the baculovirus expression system
a. Plasmids
Plasmid pV6 / 21 is a derivative of pBR322 containing a copy of the influenza A / Victoria / 3/75 virus neuraminidase gene (Van Rompuy et al., 1982). pSV 51 and both pSV 23m and pS 24m are late and early SV40-substituted sectors, respectively, and are described elsewhere (Huylenbroeck et al., 1988). pSRS-8 is a plasmid based on pPLa2311 that contains the start sequence of the hemagglutinin (NA) gene from A / Victoria / 3/75 (H3N2) (Huylenbroeck et al., 1988). The baculovirus transfer vector pVL941 was developed by Luckow and Summers (1989).

b. Subcloning of the signal peptide sequence of hemagglutinin (pIV-pRENA)
The Bst NI fragment of pSR S-8 with a size of 1830 base pairs was inserted using the Klenov enzyme. Next, Pvu II linkers (GCAGCTGC) are located. The resulting fragment and pBR322 cleave Pvu I and Pvu II and extract fragments of 731 base pairs and 1699 base pairs, respectively. Their ligation yields pIv pRENA, which contains the 5'-untranslated region of the hemagglutinin gene, starting with G ₁₆ (ATG = +1, +2, +3), followed by the intact signal sequence of hemagglutinin and the first few codons of mature hemagglutinin.

from. Construction of a chimeric sequence that encodes a secreted neuraminidase: pATIVNAS
pV 6/21 was opened with Pvu I and treated with Ba131 exonuclease. Hind III linkers are ligated to this mixture, and then Hind III is cleaved. A neuraminidase fragment of approximately 1,500 base pairs is separated and cloned into the unique Hind III site pSV 23m (pSV 23mIVNA). Plasmids with an insert in the opposite orientation then cleave Fnud and Sal I and isolate a 1291 base pair fragment containing the neuraminidase gene minus the membrane anchor sequence. Then, pIVpreNA is incubated with Pst I and Pvu II to obtain a B61br fragment with a hemagglutinin signal sequence. Both fragments are finally fused by ligation to the blunt ends of Pvu II - Fnud II and inserted into the Sal I / Pst I fragment of size 2253 base pairs from pAT153 to obtain pATIVNAs. This plasmid contains a sequence that encodes a hemagglutinin signal sequence, including the first few amino acids of mature hemagglutinin, immediately followed by a neuraminidase sequence in which there is no signal peptide / membrane anchor and part of the stem coding region. Ligation of hemagglutinin and neuraminidase fragments leads to the replacement of only one amino acid (Gly with Ala), which corresponds to position 5 in mature hemagglutinin. Based on information published by Minjou et al. (1988) and Van Rompuy et al. (1982), the predicted DNA and amino acid residue sequences flanking ligation sites in neuraminidase are shown in FIG. 2.

d. Integration of neuraminidase into baculovirus transfer vector
The XbaI / SalI fragment from pATIV NAS with a size of 1368 base pairs is ligated with a SalI / EcoRI fragment from pSV 51 of 5562 base pairs and an EcoRI / XbaI fragment from pSV24m of 624 base pairs. A copy of the 1647 base pair BamHI fragment containing the neuraminidase gene and the SV40 poly (A) site is then inserted into a single restriction site pVL941, taking into account the correct orientation with respect to the polyhedrin promoter, and pAc21VNAS is obtained. This construct provides homologous recombinant with wild-type AcNPV DNA after cotransfection of sf9 cells. The progeny of the recombinant virus is isolated by subsequent plaque purification as described by Summers and Smith (1987).

2. Insect cell culture - neuraminidase production
A typical culture of sf9 insect cells is maintained as confluent cell monolayers in TC100 medium containing 10% calf embryo serum and 50 μg / ml gentamicin. For infection with recombinant baculovirus, cultures are transferred into 200 ml suspensions, which are grown in 850 ml roller bottles (25 rpm). Then, cell suspensions at the end of their log phase (2 • 10 ⁶ cells per milliliter) are infected with recombinant baculovirus with moi ("multiplicity of infection", i.e. the number of infectious viral particles per cell) equal to 1.0. After two hours, the infected cells are transferred to a fresh portion of serum-free TC100 medium and incubated in suspension for 48 hours. Neuraminidase is purified from the culture fluid as described below.

3. Growing the X-47 influenza virus
Strain X-47 influenza is used as a source for the production of natural neuraminidase A / Victoria / 3/75 after treatment with pronase. X-47 virus is grown in the yolk sac of an 11-day-old chicken embryo. After two days in a thermostat at a temperature of 25.5 ^o C, the eggs are left to cool to 4 ^o C overnight and the yolk sac is collected for further processing.

4. Buffer systems
The following buffers are usually used:
buffer A: 20 mM diethanolamine / HCl with a pH of 8.5;
buffer B: 50 mM NaAc with a pH of 5.5;
buffer C: 10 mM NaP with a pH of 7.4, 150 mM NaCl;
Buffers A and C additionally contain 4% butanol (unless specifically stated) and 2 mM calcium chloride.

5. Purification of Recombinant Neuraminidase
a. Ammonium Sulfate Fractionation
Suspensions of sf9 culture (usually about 1 liter) are collected after infection (see above) and cell debris is pelleted by centrifugation at 4000 x g for 15 minutes. All other operations are carried out at a temperature of 4 ^o C. At the first stage of purification, a 5 mM sodium azide solution is added to the solutions. The clarified crude medium is fractionated with ammonium sulfate at pH 7.5. A substance that precipitates in the concentration range from 20% to 60% ammonium sulfate is separated by centrifugation (10,000 • g, 60 minutes) and dissolved in buffer A (without butanol) + 20 mM sodium chloride in an amount equal to one tenth of the original volume. The newly dissolved precipitate is dialyzed (with a cutoff molecular weight of 25 kilodaltons) for 24 hours against 50 volumes of the same buffer for 24 hours, with the buffer being changed three times. Insoluble components are removed by centrifugation at an acceleration of 20,000 x g for 15 minutes.

b. Anion exchange chromatography on sepharose Q
Butanol up to 4% is first added to the solution after dialysis and placed on a Sepharose Q column (2.5 cm x 10 cm), which is equilibrated with buffer A + 20 mM sodium chloride at a flow rate of 25 ml / h. After washing the column with the same buffer, elution is carried out in a linear gradient with the concentration of sodium chloride in the washing buffer to reach concentrations of 250 mM (250 ml; 25 ml / h). Fractions with a volume of 2.5 ml containing recombinant neuraminidase are identified by measuring the activity of the enzyme and its content by the method of enzyme-linked immunosorbent assay (ELISA). Neuraminidase activity elutes from the column as a single peak.

from. Affinity chromatography on agarose with N- (p-aminophenyl) oxamic acid
The use of this affinity matrix has been described for the purification of a neuraminidase (non-recombinant) influenza virus and bacterial NA enzymes (Cuatracassas and Illiano, 1971; Bucher, 1977). Correct functioning of the affinity matrix is achieved only after changing the initially recommended buffer conditions. The active fractions after separation on Sepharose Q are collected and an equal volume of 200 mM sodium acetate solution with a pH of 5.5 is added to them. Then, the active fractions are placed on an agarose column with N- (p-aminophenyl) oxamic acid (1.5 x 5 cm), equilibrated with buffer B, 100 mM NaCl. Then, the column is washed with equilibration buffer and desalted with buffer B. The second washing step is carried out using buffer A. The recombinant neuraminidase is finally eluted with buffer A containing 1 M sodium chloride at a flow rate of 10 ml / h (2 ml fractions are collected).

d. Gel chromatography on Superdex 200
The eluate obtained after the affinity column was concentrated to a volume of 2.0 ml using Centriprep ™ concentrators (from Amicon, molecular weight cut-off 30 kilodaltons). The concentrate is chromatographed in 1.0 ml fractions on a Superdex 200 gel chromatography column (1.5 x 60 cm), which is equilibrated with buffer C containing 4% butanol. The column was eluted with equilibration buffer at a flow rate of 10 ml / h, and 1.0 ml fractions were collected. For storage for a long time at a temperature of minus 20 ^o With the appropriate fractions are selected, concentrated as described previously and glycerin is added to them to a final concentration of 50%.

 To assess the molecular weight of the purified proteins, the gel chromatographic column is calibrated with horse spleen apoferritin (443 kilodaltons), sweet potato beta-amylase (200 kilodaltons), yeast alcohol dehydrogenase (150 kilodaltons), bovine serum albumin (67 kilodaltons) and 29 carbon kilodaltons) (all from Sigma Chemical Co.).

6. Preparation and purification of natural pronase-treated neuraminidase (pNA)
a. Pronase treatment
The yolk sac of chicken embryos infected with X-47 is first clarified by centrifugation at low speed (1000 • g, 10 minutes), and then centrifuged at 13000 • g for 16 hours to precipitate the virus. The virus pellet is resuspended in a volume of 10 ml of buffer C per equivalent of 100 infected eggs, and without further purification of the virus, pronase is added in an amount of 2 mg / ml. The mixture is incubated for 16 hours at a temperature of 20 ^o With slightly stirring. The remaining parts of the virus and components not dissolved by pronase are then removed in an ultracentrifuge (100,000 x g, 1 hour) at a temperature of 4 ^° C. The supernatant containing the liberated neuraminidase heads is purified by column chromatography.

b. Cation exchange chromatography on Sepharose S
Chromatography was performed at a temperature of 4 ^° C. A crude pNA sample was diluted five times and added to a 50 mM sodium acetate solution with a pH of 5.5, containing 2 mM calcium chloride and 1% butanol. Then the solution is placed on a column with Sepharose S (1.5 cm • 10 cm), which is balanced with buffer B + 1% butanol and 50 mm sodium chloride. The bound material is eluted in a linear gradient to a concentration of 500 mM sodium chloride in the same buffer. The fractions showing the highest enzyme activity were collected and concentrated to a volume of 2.0 ml using Centriprep ™ concentrator tubes (Amicon; molecular weight cut-off: 30 kilodaltons).

d. Gel Filtration on Superdex 200
Gel filtration on a Superdex 200 is carried out similarly to how neuraminidase is purified (except that the concentration of butanol is 1%). The purified pNA is stored at a temperature of minus 20 ^o C in 50% glycerol.

7. Enzymatic analysis of neuraminidase
Analysis of the catalytic activity of neuraminidase is based on the method of Potier et al. (1979). Briefly, enzyme tests are carried out in a reaction volume of 100 μl in a solution of 20 mM sodium acetate with a pH of 6.5, 2 mM calcium chloride and 1% butanol in the presence of 1 mM 2 '- (4-methylumbelliferyl) alpha-DN-acetylneuraminic acid as a substrate. After incubation at a temperature of 37 ^o C for 30 to 60 minutes, the reaction is stopped by adding 0.5 ml of 133 mm glycine, 83 mm sodium bicarbonate, 60 mm sodium chloride, pH 10.7. The content of 4-methylumbelliferone was determined by measuring the absorbance at a wavelength of 365 nm. One unit is defined as the amount of enzyme that releases one nmol of 4-methylumbelliferone per minute.

8. Immunological methods
a. Obtaining polyclonal anti-pNA IgG
Polyclonal antiserum against purified pronase treated pNA was obtained in three month old New Zealand rabbits. The primary immunization is carried out by introducing into each paw intramuscularly four doses of 500 μl, containing 50 μg of pNA / dose and 75% Freund's complete adjuvant. After six weeks, the animals receive two corresponding booster injections in both hind legs. To obtain IgG fractions, the collected serum was purified by adsorption on Protein A-Sepharose ("Pharmacia LKB").

b. ELISA
Microtiter plate cells are coated with rabbit anti-pNA IgG. Test samples are diluted with phosphate buffered saline saline containing 0.1% bovine serum albumin. The bound antigen is determined using a rabbit anti-pNA IgG treated with biotin followed by the addition of streptavidin-alkaline phosphatase conjugate ("Boehrinqer"). An enzymatic reaction develops when the plates with p-nitrophenyl phosphate are incubated (Sigma Chemical Co.). Absorbance values are determined at a wavelength of 405 nm using a microplate reader.

9. Analytical methods
Polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate was carried out according to the method of Laemmli (1970) on a 10% separating gel (unless specifically indicated). All specimens, unless specifically indicated, are denatured in the presence of beta-mercaptoethanol. As marker proteins in a 10% gel, phosphorylase b (94 kilodaltons), bovine serum albumin (67 kilodaltons), ovalbumin (43 kilodaltons), carbonic anhydrase (29 kilodaltons) and a trypsin inhibitor (20.1 kilodaltons) are not always visible ) ("Pharmacia LKB"). Gradient gels are calibrated with the following mass standards: myosin (22 kilodaltons), beta-galactosidase (116 kilodaltons), phosphorylase b, bovine serum albumin and ovalbumin ("Biorad"). Staining with silver in gels is carried out according to a modified method described by Morrisey (1981). Protein concentration is determined according to the method of Bradford (1976), using ovalbumin as a standard.

10. Analysis by cross-linking
The BS ³ molecule for cross-linking is used as a freshly prepared 1.0 M solution in 10 mM Hepes buffer, pH 7.4. Proteins are cross-linked by the addition of BS ³ to a concentration of 0.5 mM in a reaction volume of 30 μl. Incubation is carried out for 1 h at room temperature. Then the reaction is interrupted by adding 5 μl of 1.0 M Tris buffer with a pH of 8.0. Polypeptide samples are analyzed by polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate.

11. Carbohydrate analysis
Protein samples (0.1 μg to 1 μg) were denatured by boiling in 500 mM Tris / Hcl with pH 8.0, 0.5% sodium dodecyl sulfate, 50 mM beta-mercaptoethanol. After the addition of neuraminidase-octyl glucoside to a concentration of 2.5%, which leads to at least a seven-fold increase over the final concentration of sodium dodecyl sulfate, neuraminidase-glucanase is added (about 0.5 units; the number of units is indicated according to the manufacturer) and the reaction mixture is incubated at a temperature of 37 ^o C for 16 hours. Cleavage patterns are analyzed by polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate.

results
1. Purification of pNA
In the typical experiment described here, 186 infected eggs are processed. The various stages of purification are summarized in table. 1. After collecting the liquid from the yolk sacs and sedimenting the virus, pronase is added to a concentration of 2 mg / ml and the mixture is incubated at a temperature of 20 ^{° C.} for 16 hours. After ultracentrifugation in liquid above the pellet, approximately 60% of the neuraminidase activity is determined. It was shown that under these conditions, the loss of activity is mainly associated with incomplete extraction of the heads of the neuraminidase of viral particles. Higher pronase concentrations, longer incubation periods or higher temperatures do not lead to an increase in yield, since neuraminidase is gradually degrading (data not shown). The pNA raw material is then dissolved, adjusted to pH 5.5 and placed on a cation exchange column with Sepharose S. For maximum pNA output, all subsequent solutions contain 1% butanol. Most of the protein is not retained in the Sepharose S column, and after gradient elution, a single peak (not shown) is collected at a sodium chloride concentration of approximately 400 mM. This substance contains substantially purified pNA: since no contaminating bands were detected after polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate (Fig. 3A, lane 3). Further, silver staining does not show any differences between the pool after Sepharose S and after additional gel filtration on Superdex 200 (Fig. 3A, lanes 3 and 4). After the last column, a single bell peak is collected in fraction 60 (not shown), which corresponds to a molecular weight of approximately 210 kilodaltons. Successive cleaning steps are illustrated in FIG. 3A.

During polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate, the pNA monomer is actually visible as a doublet of two bands, corresponding to approximately 54 kilodaltons and 52 kilodaltons, respectively, the latter being most common, as evidenced by the intensity when stained with silver. Most likely, this ambivalence appears due to the preferred cleavage by pronase in two different places of the stem region. Cross-linking with the chemical agent BS ³ confirms that pNA is isolated as an authentic tetrameric protein (Fig. 3B).

2. Construction and expression of recombinant neuraminidase
The neuraminidase gene of strain A / Victoria / 3/75 of influenza NA2 is separated from its terminal membrane anchor of neuraminidase and attached instead to the 5'-sequence of the hemagglutinin gene A / Victoria / 3/75, which contains the signal peptide splice site. It becomes possible to synthesize a secreted soluble product. The resulting chimeric gene contains a hemagglutinin signal sequence including codons for the first four terminal amino acids of mature hemagglutinin, immediately followed by a neuraminidase sequence in which there is no transmembrane part (anchor) and part of the stem region (amino acids 1 to 45). Both DNA sequences are located in the same reading frame, with no additional amino acids being introduced. Ligation leads to only one amino acid change corresponding to position 5 of the mature hemagglutinin protein (Fig. 2). A copy of this chimeric sequence, which now mainly encodes a secreted protein, is integrated behind the AcNPV baculovirus polyhedrin promoter using pVL941 as a transfer vector. After infection of the insect sf9 cells, neuraminidase activity in this medium is quickly determined and it is shown that it actually obtains a soluble protein.

 From FIG. Figure 4 shows that the activity of neuraminidase in the medium reaches a plateau level approximately 48 hours after infection. Further incubation does not bring results, since the total concentration of the protein begins to decrease sharply, probably due to excessive cell lysis. It was shown that expression most intensively occurs when the intermediate transfer between the initial sf9 monolayer and the suspension culture is limited to a minimum (data not shown). Based on a variety of purification experiments, it was found that the expression level of neuraminidase varies from 6 to 8 mg / L, which corresponds to a rather low productivity, however, it is comparable to the yields reported for other secreted complex glycoproteins produced in this system (Jarvis et al ., 1990).

3. Purification of Recombinant Neuraminidase
TC100 medium is collected 48 hours after infection at the moment when the specific enzymatic activity of the soluble protein reaches a peak value (Fig. 4). The various stages of purification of recombinant neuraminidase are summarized in table. 1. The precipitation of the crude medium with ammonium sulfate in the saturation range from 20 to 60% gives an average double enrichment and allows the concentration of the product. After intensive dialysis and removal of insoluble products, butanol is added to a concentration of 4%. The addition of butanol has been shown to have a significant beneficial effect on the overall yield of recombinant neuraminidase, especially at low protein concentrations. Perhaps, to prevent the formation of insoluble aggregates, a certain hydrophobicity of the medium is necessary. Then, the solution was fractionated by anion exchange chromatography on Sepharose Q (Fig. 5). The activity of neuraminidase elutes at the beginning of the salt gradient in the form of an almost symmetrical peak. According to the solid-phase immunosorbent assay, the remaining fractions do not contain neuraminidase-related substances. At this stage, approximately 97.5% of the initial amount of protein is isolated, which leads to an increase in specific activity by approximately 20 times. Then the pH of the solution before loading into a column with N- (p-aminophenyl) oxamic acid - agarose is reduced to a value of 5.5. From early studies, it is known that neuraminidase-substituted oxamic acids are potent reversible inhibitors of influenza neuraminidase (Edmond et al., 1966). The use of N- (p-aminophenyl) oxamic acid agarose as a selective absorbent of influenza neuraminidase or bacteria was first described by Cuatrecasas and Illiano (1971) and later by Bucher (1977). In accordance with the original method, neuraminidase is eluted with a high pH buffer (100 mM sodium bicarbonate, pH 9.1). However, in our experiments, these conditions allow only partial and slow extraction of recombinant neuraminidase. However, effective desorption can be achieved by combining an increase in pH with a high salt concentration. Before elution, a significant amount of non-specifically bound protein is removed from the column by performing an additional washing step at pH 8.5 with a low salt concentration. If diethanolamine is preferred as a buffering agent over sodium bicarbonate, absorption of 2 mM calcium chloride can be achieved without precipitation. It was repeatedly indicated that the preservation of neuraminidase activity is somewhat dependent on the concentration of calcium ions (Chond et al., 1966; Dimmock, 1971). Is this so, this study does not analyze in detail.

To remove traces of residual contaminants, the eluate is concentrated by ultrafiltration and subjected to gel filtration on a Superdex 200 (Fig. 6). The A ₂₈₀ test shows three peaks with unequal absorption that elute, respectively, at about 220 kilodaltons, about 130 kilodaltons, and about 54 kilodaltons. It was shown that the eluate immunoreactivity patterns measured by enzyme-linked immunosorbent assay are a reliable representation of the A ₂₈₀ profiles for each of the three registered peaks, which indicates that all compounds are NAs-specific. Analysis of peak fractions by polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate shows an intense band in the expected region of approximately 55 kilodaltons, although a slight decrease in molecular weight is observed with increasing fraction number (Fig. 7A). A peak of 220 kilodaltons in cross-linking assay using BS ^{3 is} identified as a tetrameric recombinant neuraminidase, and the other two peaks with a lower molecular weight are found to be dimeric and monomeric neuraminidase, respectively, while the latter form is of negligible quantitative significance (Fig. 7B). It is believed that due to its rod-like structure, the dimeric recombinant neuraminidase elutes slightly higher than its molecular weight compared to the tetrameric and monomeric recombinant neuraminidase, which are believed to have a circular shape. Of particular note, catalytic activity requires a fully assembled tetrameric structure of the recombinant neuraminidase. Perhaps the formation of the tetramer causes several local conformational changes, which are important for enzymatic activity.

 The sequence of cleaning operations is given in table. 2.

4. Properties of recombinant neuraminidase
Denaturation with a boiling solution of sodium dodecyl sulfate in the presence of beta-mercaptoethanol causes complete dissociation of the recombinant neuraminidase into monomeric chains with a molecular weight close to 55 kilodaltons (Fig. 7A). The tetrameric and dimeric recombinant neuraminidase are homogenously purified, judging by the silver staining of the gels by polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate. Monomeric recombinant neuraminidase has a slightly worse quality, since several traces of contaminating compounds are observed. When they are denatured in the absence of a reducing agent, the tetrameric and dimeric recombinant neuraminidase migrate in the form of dimeric chains of approximately 110 kilodaltons (not shown). The results obtained indicate that the recombinant neuraminidase dimers actually bind with disulfide bridges and can be further associated via non-covalent interactions, forming a tetramer that corresponds to the structural organization of natural neuraminidase.

It has been repeatedly reported that insect cells provide a glycosylation pattern of neuraminidase that differs to some extent from that of mammalian cells and other cells of higher animals (Hsieh and Robbins, 1984; Butter and Hughes 1981; Butters et al., 1981; Kiroda et al. , 1990). In this regard, the amount of neuraminidase-bound carbohydrate bound recombinant neuraminidase is compared with natural pronase-treated neuraminidase. Individual protein samples are treated with the enzyme NA-glucanase, and then analyzed by polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate (Fig. 9). Based on the relative shift of the bands, it can be concluded that the total amount of NA-bound carbohydrate by recombinant neuraminidase is slightly less than the natural molecule (compare Figs. 9A and 9B), and this observation is consistent with the observation for other glycoproteins expressed in this system (Kiroda et al., 1986; Domingo and Trowbridge, 1988; van Drunen Littel et al., 1991). It was also found that denatured enzymatically deglycosylated recombinant neuraminidase forms migrants with the same electrophoretic mobility regardless of their initial oligomeric structure, and this confirms the fact that the primary recombinant neuraminidase is synthesized as a polypeptide with a uniform chain length (Fig. 9B, compare lanes 3, 5 and 9). The molecular weight of the polypeptide chain after treatment with NA-glucanase is estimated at 47.5 kilodaltons, which corresponds to a theoretical value of 47,717 daltons based on the predicted sequence of amino acid residues. It is interesting enough to note that the degree of glycosylation of neuraminidase is probably associated with the ability to form tetramers, since the bands corresponding to the glycosylated dimeric and monomeric neuraminidase move slightly faster in the gel than the bands of glycosylated tetrameric neuraminidase (Fig. 9B, lanes 4 and 6 compared to lane 2 ; see also Fig. 7A). Indeed, it has been suggested that a hydrocarbon bound to a neuraminidase, namely an oligosaccharide chain attached to Asn ₂₀₀ , may play a role in establishing a tetrameric structure due to interaction with adjacent subunits (Varghese et al., 1983; Varghese et Colman, 1991).

 Only tetrameric protein contributes to the catalytic properties of the recombinant neuraminidase. The isolated tetrameric recombinant neuraminidase has a specific activity that is almost identical to the specific activity of purified natural neuraminidase treated with pronase (Tables 1 and 2). The observation that the form of the recombinant lower structural order neuraminidase is enzymatically inactive, even though each monomer has a catalytic cavity, probably reflects the role of the structure in the Quaternary interactions during the functioning of influenza neuraminidase.

In order to verify the antigenic properties of the recombinant neuraminidase, protein samples of equal concentration are diluted twice in succession and tested in a sandwich immunoassay enzyme-linked immunosorbent assay using polyclonal anti-pNA IgG (Fig. 10). The titration curve of the tetrameric recombinant neuraminidase is identical to the comparison curve of natural pronase-treated neuraminidase, indicating that they both have identical or very similar antigenic properties. Despite the absence of noticeable enzymatic activity, the antigenic activity of the dimeric and monomeric recombinant neuraminidase remains almost unchanged, although a slight shift in antigenicity may be observed. This slight difference in antigenicity is also evident from the gel filtration profile (Fig. 6), where the antigenic activity / A ₂₈₀ ratio shows some superiority of the tetramer peak. It is possible that some antigen molecules generated against the natural tetrameric structure are not able to efficiently bind to incompletely assembled recombinant neuraminidase, for example, those molecules that recognize contact regions between neighboring subunits. Local conformational changes caused by the formation of the tetramer can also lead to slight antigenic differences.

The discussion of the results
The main objective of the present invention is the synthesis of influenza neuraminidase antigen in the form of a secreted and properly folded protein, together with the choice of purification procedure in order to obtain a homogeneous product that can be used as a means for vaccination. From the neuraminidase gene of strain A / Victoria / 3/75 of influenza NA2, a chimeric gene is constructed in which the initial NA-terminal region, which has a combined function of the signal sequence,
the membrane anchor is replaced by part of the 5 ′ sequence of the influenza neuraminidase gene. The resulting genetic engineering construct, due to the cleavable signal peptide derived from hemagglutinin and encoding mainly secreted neuraminidase, is then incorporated into the expression vector of baculovirus under transcriptional regulation by the potent polyhedrin promoter. After infection of the sf9 host cell, recombinant neuraminidase is indeed secreted into the culture medium. Based on the results of the purification, the expression level is estimated at 6-8 mg / L. It has been shown that in the process of baculovirus infection, the ability of the host cell to produce proteins by secretion is significantly reduced (Jarvis and Summers, 1989). Nevertheless, the described production system is applicable for the study of vaccinations in the laboratory and is suitable for use on a large scale.

 Purification of recombinant neuraminidase mainly consists of a four-step process, the first stage of which is fractionation with ammonium sulfate, followed by three successive stages of chromatography. Based on the yield of enzymatic activity, the amount of recombinant neuraminidase isolated as a protein is estimated at approximately 25%. Using chromatography on a gel chromatographic column, recombinant neuraminidase is subfractionated into three populations with different molecular sizes, which are identified by cross-linking, respectively, as tetrameric, dimeric, and monomeric recombinant neuraminidase, the latter form being present only in very small quantities. The two main forms, tetrameric and dimeric recombinant neuraminidase, are obtained in approximately equal amounts, as shown by polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate, followed by silver staining.

 To evaluate the enzymatic and immunological properties of the recombinant neuraminidase, it is necessary to isolate the natural pronase-treated neuraminidase as a reference protein. The head of the Neuraminidase A / Victoria / 3-75 of the X-47 virus is cleaved by treatment with pronase, and then purified by cation exchange and gel chromatography. Cross-linking confirms that recombinant neuraminidase retains the tetrameric structure of an intact neuraminidase attached to the membrane.

 The catalytic properties of recombinant neuraminidase are very remarkable, since only tetrameric protein exhibits enzymatic activity. The specific activity of tetrameric recombinant neuraminidase is almost equal to the specific activity of natural pronase-treated neuraminidase. It is likely that the dimeric and monomeric recombinant neuraminidase are inactive due to the fact that they are denatured proteins, because during the purification process, these forms were firmly retained by affinity chromatography at the sites of binding of the sustrate, indicating that the enzymatic cavity was not functionally affected, however, the subsequent catalytic shift not to be observed.

 Treatment with NA-glyconase shows that, in general, the carbohydrate content in recombinant neuraminidase is slightly lower than in natural pronase-treated neuraminidase, which is also observed for other glycoproteins expressed in this system (Kuroda et al., 1986; Domingo and Trowbridge, 1986; van Drunen Littel et al., 1991). Hypoglycosylation is apparently more pronounced for dimeric and monomeric recombinant neuraminidase.

Structural studies using x-ray diffraction analysis show that the carbohydrate chain attached to A ₂₀₀ is in close contact with neighboring subunits, suggesting that this may contribute to the emergence of additional interactions that strengthen the quaternary structure (Vaqhese et al., 1983; Vaghese and Colman, 1991).

 The reactivity of tetrameric recombinant neuraminidase with respect to polyclonal immunoglobulins generated against purified natural pronase-treated neuraminidase is almost complete, indicating that both proteins have very similar antigenic properties. A slight shift in antigenicity can be observed in the case of dimeric and monomeric recombinant neuraminidase. It follows that it is probably possible to isolate monoclonal antibodies that bind only to the tetrameric structure of influenza neuraminidase. Such antibodies are likely to interact with surface determinants formed by neighboring subunits, or they are able to recognize epitopes that arise after conformational rearrangements during the formation of the tetramer. In addition, differences in carbohydrate composition can also modulate antigenic properties.

Example 2. The secretion of recombinant neuraminidase Pichia pastoris
Introduction
In order to find out whether yeast can be used in addition to insect cells to produce recombinant influenza neuraminidase, an expression vector is constructed that contains the “cap” of the neuraminidase with enzymatic activity.

Materials and methods
1. Vector and host
To construct a polygenic expression cluster, plasmid pIPC9 from Pichia pastoris (Invitrogen) was used. This plasmid contains the replication initiation site, the ampicillin resistance gene, the promoter region and the terminator of the introduced alcohol oxidase 1 (AOX1) gene from P. pastoris, the prepro signal for the secretion of the alpha factor Saccharomyces cerevisiae, and the HIS4 marker from P. pastoris.

 As the host, methylotrophic yeast Pichia pastoris ("Invitrogen") is used.

2. Construction of a polygenic expression cluster
By the method of site-oriented mutagenesis, the Stu I restriction site is introduced into the cDNA sequence of the neuraminidase gene A / Victoria / 3/75. The restriction site is located at position Pro79. Using this restriction site, one can isolate the immunogenic hat sequence of the neuraminidase gene containing the active site as a Stu I / Hind III fragment, and clone it at the Sna BI restriction site of the plasmid pIPC9 from P. pastoris. In FIG. 15 is a diagram of plasmid pIPC9. In FIG. Figure 16 shows a view of the fusion region between the prepro signal sequence and recombinant neuraminidase. The propeptide is cleaved by the Golgi method by the endogenous KEX2 protease. The dipeptide (Glu-Ala) _{2 is} isolated using the STE13 type dipeptidyl aminopeptidase. The excess tyrosine residue is not cleaved and stored in the recombinant neuraminidase as an N-terminal residue, but it is not needed.

 The resulting plasmid is linearized at the position of the HIS 4 marker of selection by cleavage with Sall, and then transformed into GTS115 (his 4) P. pastoris protoplasts in the presence of polyethylene glycol. DNA isolated from transformed cells is analyzed by southern blotting. The analysis showed that the expression vector is integrated by homologous recombination into the position of the internal (but incomplete) locus of his 4. Most of the transformed cells contain from 1 to 2 copies of the plasmid, however, it was found that the transformed cells with greater secretory ability contain several copies that integrate head-to-tail type in the host genome in a tandem structure. The number of copies can reach up to 25 per transformed cell.

3. Expression of neuraminidase
Transformed cells are pre-grown in a minimum amount of glycerol containing a buffer (pH 6.0) and after 48 hours they are transferred to a minimum amount of a buffer solution containing 0.5% methanol. The alcohol oxidase 1 promoter is introduced and a “cap” of neuraminidase is expressed. Using the well-known Northern blot analysis method, the amount of neuraminidase mRNA in the cell is estimated. Analysis shows that a very effective induction has occurred.

 Western blot analysis of the liquid over the sediment shows that recombinant neuraminidase with a molecular weight of approximately 70 kDa is expressed (see FIG. 17).

 The secreted product is subjected to deglycosylation with PNase F. An “core” product is obtained with an expected size of 43 kDa. It was found that, depending on the number of copies, the yield of recombinant neuraminidase in the medium varies from 1 to 1.5 mg / L.

Example 3. Immunization
Materials and methods
1. Animals
The age of female inbred individuals of Balb / c mice (from the company SCK Mol, Belgium) at the beginning of the immunization was 8 weeks. When conducting passive immunization, the age of the recipient mice was 12 weeks. Mice are divided into groups of three animals per cage (410 cc) and are given plenty of food and water.

2. Viruses
Virus strains were obtained from Dr. A. Douglas and Dr. J. Skehel (MCR Laboratories, Mill Hill, London). Laboratory viruses X-31 and X-47 have an H3N2 antigenic composition and are obtained by genetic rearrangement from A / PR / 8 // 34 (H1N1), respectively, with A / Aishi / 2/68 (H3N2) and A / Victoria / 3 -75 (H3N2). Both sources of viruses are adapted by several re-grafts into the lungs so that they cause death in mice.

3. Secreted recombinant neuraminidase
Influenza neuraminidase A / Victoria / 3/75 is administered as a purified recombinant protein produced using an insect baculovirus-cell expression system as described in Example 1. The purified recombinant neuraminidase preparations used for the described immunization experiments contain a mixture of tetrameric and dimeric molecules in buffered phosphate saline.

4. Adjuvants
Suitable adjuvants are selected based on immunization studies using recombinant influenza hemagglutinin in our laboratory. Ribi adjuvant (with monophosphoryl lipid A (MPLA), trehalose-6,6-dimethicolate (TMD), squalene and Tween 80) and Salmonella typhimurium MPLA are filled in bottles according to the manufacturer's recommendations (Ribi Immunochem Reсeca). Muramyldipeptide (MDR) is purchased from Sigma Chemical Co.
5. Immunization Technique
Three doses of 200 μl, each containing 1 μg of recombinant neuraminidase, are subcutaneously injected into mice with a three-week interval. For the first immunization, recombinant neuraminidase is emulsified in half the usual dose of Mouse Ribi (corresponding to 25 μg MPLA, 25 μg TMD, 2 μl squalene and 0.1% Tween 80). Active immunizations are carried out by adding 25 μg of MPLA and 25 μg of MDP to recombinant neuraminidase. Control animals receive an adjuvant dissolved in phosphate buffered saline.

6. Passive immunization
Three weeks after the third immunization, blood samples were taken from donor mice by heart puncture and the serum preparations from the respective vaccinated mice were combined. Recipient mice receive an intraperitoneal injection of 400 μl of immune or control serum.

7. Infection with influenza
Under mild ether anesthesia, mice are infected with a dose of 20 LD _{50 of a} particular virus three weeks after the last active immunization or one day after passive immunization. Infection is monitored for 10 days by measuring rectal temperature and body weight after vaccination.

8. Serological method
One day before the start of the vaccination procedure (pre-immune serum) and two weeks after each immunization, blood samples are taken from the caudal artery. Individual serum samples are tested for NA-specific antibodies using enzyme-linked immunosorbent assay. The titration plates ("Nunc Maxisorp") are coated with purified recombinant neuraminidase (50 ng per well) and the serum is diluted 1: 5. The binding of a specific antibody is quantified by adding rabbit antibodies to anti- (mouse IgG) conjugated to alkaline phosphatase (Sigma Chemical Co.), and then the plates are incubated with a solution of p-nitrophenol (from Sigma Chemical Co. "), which is a substrate. The optical density is determined at a wavelength of 405 nm using a reader tablets for microtiter. The neuraminidase antigen titer is expressed as the reciprocal of the log _s dilution of serum and an absorption value is obtained that is 0.05 higher than in control cells (treated with pre-immune serum).

results
1. Research procedure
Three groups of 12 mice are vaccinated with recombinant neuraminidase in accordance with the above immunization procedure. An equal number of control mice are simultaneously treated with phosphate buffered saline. Paired groups of vaccinated and control mice are then infected with X-47 and X-31 adapted for mice. In an alternative method, they serve as serum donors for passive immunization experiments.

2. Serological response
Responses of antigens against recombinant neuraminidase are monitored by enzyme-linked immunosorbent assay in the serum of randomly selected 12 vaccinated and 12 control animals (one animal from each cell) (Fig. 11). The use of recombinant neuraminidase for immunization in mice causes a steady increase in the content of antibodies to recombinant neuraminidase in serum. The first active immunization causes an increase in the amount of antibodies in an amount of about three log _s, while reimmunizatsiya leads to another approximately five-fold increase in antibody titer to the recombinant neuraminidase. In control mice, injection of an adjuvant alone does not significantly produce specific antibodies that interact with recombinant neuraminidase.

3. Homovariant protection using recombinant neuraminidase
Three weeks after vaccination, immunized mice and control mice were tested for immune activity by administering a dose of 20 LD _{50 of the} homo-NA variant X-47 virus (FIG. 12). All control animals became seriously ill, which is determined by measuring the increasing drop in body temperature and by the loss of body weight. On the fourth day after infection, the first victims are detected, and after nine days after infection, all control animals died. In contrast, in animals that were given recombinant neuraminidase vaccines, only a temporary and very moderate change in clinical parameters was observed. All immunized animals suffered an infection.

 It is also being examined whether the same level of protective immunity can be obtained without a third immunization simply by applying compensation in the form of large doses of recombinant neuraminidase and / or adjuvant. These tests are carried out according to the same scheme. Although mice vaccinated in this way usually show good resistance, several individuals become seriously ill, and some vaccinated animals eventually die, although the percentage of surviving animals is rarely less than 80%. However, it was found that the level of protective immunity achieved through three vaccinations exceeds in all respects the level of protection after two vaccinations.

4. Heterovariant protection using recombinant neuraminidase
Parallel groups of vaccinated and control mice were tested for immunity using a dose of 20 LD ₅₀ hetero-NA-variant virus X-31, which contains neuraminidase isolated from recombinant neuraminidase from A / Victoria / 3/75 after 7 years of antigenic drift (Fig. 13) . During immunization using X-47, we also observed that the clinical consequences of infection in the control group were dramatic. Control animals began to die as early as five days after infection. Mortality peaked on the eighth day, when only one animal remained. Mice injected with recombinant neuraminidase vaccines show 100% survival after commonly fatal heterovariant infections. In the same way as in the study of homovariant immunization, vaccinated animals were also able to maintain their body temperature at a reasonably normal level. Loss of body weight is somewhat brighter, but all mice begin to recover on the sixth day.

 5. Protective immunity can be achieved by passive transfer of the NAS immune serum.

To determine whether humoral defense mechanisms, which are primarily responsible for the induction of protective immunity, are relevant, animal protection is tested by passive immunization. To do this, donor mice are vaccinated according to a standard method. Blood is taken and approximately 400 μl of serum is obtained from each animal. Combine, on the one hand, samples of control serum, and on the other hand, samples of immune serum and recipient mice are injected intraperitoneally with one dose of serum in an amount of 400 μl. Before infection with a dose of adapted X-47 virus in an amount of 20 LD ₅₀ wait for 24 hours so that the antibody molecule systemically spread in the body of the mouse. While mice that received control doses of serum later developed acute hypothermia and rapid weight loss, which ultimately leads to death, the administration of immune serum containing recombinant neuraminidase protects the mice in much the same way as in the case of active vaccination (Fig. 14). From this we can conclude that the presence of already circulating NA antibodies is sufficient to provide complete protection.

The discussion of the results
Immunity to influenza for a long time was studied almost exclusively as a function of the hemagglutinin antibody, while the important contribution to the immunity of neuraminidase was largely ignored. This is partially explained by the observation that only antibodies capable of binding hemagglutinin were able to directly neutralize virus particles (Nirst, 1942; Davanport et al., 1964; Kida et al., 1983), while antibodies against neuraminidase concentrations obviously could not prevent primary infection (Jahiel and Kilbourne, 1966; Kilbourne et al., 1968; Johanson, 1989). This tolerance probably reflects the recently identified role that recombinant neuraminidase plays in the life cycle of the influenza virus, preventing the newly formed virus from aggregating on the surface of the infected cell (Colman and Ward, 1985; Brown and Laver, 1968). Moreover, it was found that, in contrast to hemagglutinin, neuraminidase forms a smaller part of the influenza virus envelope and this fact may also contribute to the inability of the neuraminidase antibodies to have a neutralizing effect (Schulman et al., 1968). This difference in molar presence also affects relative antibody responses to individual antigens. The repeated dominance of hemagglutinin over neuraminidase due to a successful confrontation with the entire influenza virus could lead to a suppression of the production of neuraminidase antibodies, probably due to the weakened help provided by neuraminidase-specific T cells (Kilbourne, 1976; Johanson et al., 1987; Kilbourne et al., 1987; Kilbourne et al., 1987; Kilbourne et al., 1987; Kilbourne et al., 1987; Kilbourne et al., 1987 al., 1987; Johanson et al., 1987). In order to study the protective immunity of neuraminidase, it is necessary to develop a system in which the interfering effect of antibodies neutralizing hemagglutinin is eliminated, and the suppression of the immune response of neuraminidase due to competition between hemagglutinin and neuraminidase antigens is avoided. Classical approaches were based either on the isolation of a component of natural neuraminidase (Schulman et al., 1968; Johanson and Kilbourne, 1990; Gallagher et al., 1984), or on the combined use of certain series of influenza strain with selologically different hemagglutinin and neuraminidase antigens (Rott et al., 1974; Kilbourne, 1976). The results obtained in this work directly demonstrate protective immunization using purified recombinant neuraminidase protein. The A / Victoria / 3/75 (H3N2) virus neuraminidase gene is transformed into a gene that encodes a secreted protein (recombinant neuraminidase) by replacing the region encoding the membrane anchor with the signal sequence of the influenza hemagglutinin gene (see Example 1).

 It has already been established by in vitro methods that neuraminidase antibodies can effectively suppress the increase in virus output by inhibiting the release and spread of virus particles (Jahiel and Kilbourne, 1968; Kilbourne et al., 1968). Similar conclusions were made for animals immunized with neuraminidase when measuring reduced titers of the virus in the lungs and reducing the likelihood of developing lung lesions (11, 12, 13).

 Although considerable attention was paid to the effect of NA immunity on the replication of the virus in the lungs, the question remained whether immunization with pure neuraminidase protein could prevent the clinical symptoms of the disease or could increase the chance of survival after a potentially fatal influenza infection. This question has not yet been answered satisfactorily. The results presented in the present description clearly show that by immunization with pure recombinant neuraminidase, complete protection against a usually lethal dose of infection can be achieved, while any possible contribution from the anamnestic anti-hemagglutinin immune mechanism or the effects of cell-mediated immune memory against antigens of conserved internal viral proteins.

 During the experiments described herein, mice are immunized with three doses of 1 μg of recombinant neuraminidase, which is prescribed at three-week intervals. Vaccinated animals are able to fully survive a lethal infection of the influenza virus, while the virus expresses homo- and heterovariant recombinant neuraminidase. In view of the high doses of the virus infection, it was very surprising that in animals after a well-conducted immunization there are no disease symptoms that are determined by changes in body temperature and weight. It is important to note that the adjuvants that are prescribed together with recombinant neuraminidase all have low reactogenic properties, so that the immunization procedure described in this description can be directly used to vaccinate people. The vaccines of the present invention can also be used for other mammals and birds.

 Passive transfer of the serum of mice immunized with recombinant neuraminidase to recipient mice leads to the same level of protection, indicating that the protective effect of immunization with recombinant neuraminidase can be explained by circulating neuraminidase antigens.

 Regarding the heterovariant protection discussed in the present description, it is important to consider the structural relationships between the neuraminidase antigen of vaccine A / Victoria / 3/75 and the neuraminidase from A / Aichi / 2/68, which is present in the variant of the infectious virus X-31. Unfortunately, data on the sequence of amino acid residues for neuraminidase from A / Aichi / 2/68 (H3N2) were not obtained, however, a comparison can be made with the sequence of amino acid residues of neuraminidase from strain A / NT / 60/68 (H3N2) (Bentley and Brownley, 1982), isolated in the same year as the Aichi strain. A careful study of the head region of both variants of neuraminidase reveals amino acid substitutions at positions 28, with most of them located on the surface of the molecule.

 Probably, the vaccine of the present invention can also provide protection against even more remote due to drift options. It is also possible that with the help of gene modifications of the neuraminidase gene, variants can be arranged in its antigenic structure. In this case, it becomes possible, for example, to obtain a “cocktail” of various versions of neuraminidase and to achieve intensive protection against various strains of influenza.

Figures
In FIG. Figure 1 shows a strategy for constructing a secreted neuraminidase gene and integrating it into a baculovirus transfer vector. Only the required restriction sites are indicated. Thin lines indicate bacterial plasmid sequences, while bold lines indicate hemagglutinin-specific (shaded) and neuraminidase-specific (dotted) sequences. The signal sequence of hemagglutinin is indicated by a single hatch. The neuraminidase signal sequence / membrane anchor sequence is indicated by double hatching.

 In FIG. 2 shows the nucleotide sequence of the positive branch of cDNA and the sequence of amino acid residues of the flanking regions of the ligation site between the hemagglutinin and neuraminidase signal peptide, while the membrane anchor of the neuraminidase is removed.

In FIG. 2A shows unprocessed hemagglutinin and a detailed signal peptidase restriction site between Ala ₁₆ and Gln ₁₇ (vertical dashed line). The terminal segment of hemagglutinin, which is used for the secretion of recombinant neuraminidase, is indicated by an arrow.

 In FIG. 2B shows in detail the "stem" region of neuraminidase. The truncated sequence involved in the construction of recombinant neuraminidase is indicated by an arrow.

 In FIG. 2C shows how a recombinant neuramidinase sequence is constructed from A and B. The fusion region between the hemagglutinin-specific and neuraminidase-specific sequences is shown in detail. Recombinant neuraminidase appears to start with four terminal amino acids of neuraminidase derived from mature hemagglutinin, followed by a mutated codon (underlined by a dashed line).

In FIG. Figure 3 shows the result of analysis of purified natural pronase-treated neuraminidase by polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate. FIG. 3A relates to the analysis of protein samples taken at various stages during the purification of natural pronase-treated neuraminidase. Lane 1 shows marker proteins; lane 2 corresponds to the raw material of natural pronase-treated neuraminidase (1 μg; strip of natural pronase-treated neuraminidase below the detection limit); lane 3 shows the combined eluate after Sepharose S (1 μg); lane shows pooled eluate after Superdex 200 (1 μg). In FIG. 3B shows a gradient from 5.0% to 7.5% gel of 1 μg of natural pronase-treated neuraminidase, which is cross-linked using BS ³ . Lane 1 corresponds to marker proteins; lane 2 corresponds to natural pronase-treated neuraminidase after cross-linking. Additional bands appeared at about 105 kDa (dimer), about 160 kDa (trimer), and about 210 kDa (tetramer).

обнаруженной в среде для выращивания культуры, которую оценивают из уровней ферментной активности

и общей концентрации белка

после заражения клеток sf9 рекомбинантным бакуловирусом.In FIG. Figure 4 shows the change in specific activity over time.

found in the culture medium, which is evaluated from the levels of enzyme activity

and total protein concentration

after infection of sf9 cells with recombinant baculovirus.

и антигенность иммуноферментным твердофазным анализом

На фиг. 6 приведен результат гель-фильтрации рекомбинантной нейраминидазы на Superdex 200. Элюат после разделения на N-(п-аминофенил)оксамовая кислота - агароза (2,63 мг белка, 49100 единиц) концентрируют до объема 2,0 мл и затем хроматографируют, собирая фракции по 1,0 мл, на колонке с Supеrdеx 200 со скоростью подачи 10 мл/ч. Непрерывно проводят контроль значения A₂₈₀(_ _ _ __). Индивидуальные фракции (1,0 мл) испытывают на ферментативную

а также антигенную активность

Стрелки показывают объем элюции калибровочных белков (см. текст): 443 кДа (1), 200 кДа (2), 150 кДа (3), 67 кДа (4) и 29 кДа (5).In FIG. Figure 5 shows the result of anion-exchange chromatography in a column with Sepharose Q. After dissolution and dialysis of the precipitate obtained with (20-60)% sodium sulfate, a solution (97.5 mg of protein, 117,000 units) is placed in a column with Sepharose Q. Unbound substances are washed out before elution, which is carried out in a gradient of sodium chloride (---) to a concentration of 250 mm, added to the initial buffer solution. After protein concentration in the eluate, A ₂₈₀ (_____) measurements are taken. 2.5 ml fractions were collected and enzymatic activity was studied.

and antigenicity by enzyme-linked immunosorbent assay

In FIG. Figure 6 shows the result of gel filtration of recombinant neuraminidase on Superdex 200. After separation into N- (p-aminophenyl) oxamic acid - agarose (2.63 mg protein, 49100 units), the eluate was concentrated to a volume of 2.0 ml and then chromatographed to collect fractions 1.0 ml, on a column with Supеrеx 200 with a feed rate of 10 ml / h. Continuously monitor the value of A ₂₈₀ (_ _ _ __). Individual fractions (1.0 ml) were tested for enzymatic

as well as antigenic activity

The arrows show the elution volume of the calibration proteins (see text): 443 kDa (1), 200 kDa (2), 150 kDa (3), 67 kDa (4) and 29 kDa (5).

In FIG. 7 shows the result of analysis by polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate of purified natural pronase-treated neuraminidase. Each lane corresponds to a specific fraction number from a Superdex 200 column after gel filtration. In FIG. 7A shows a polyacrylamide gel electrophoresis pattern in the presence of sodium dodecyl sulfate after denaturation of 10 μl of the sample (beta-mercaptoethanol is added). Lanes A and B correspond to marker proteins. In FIG. 7B, protein samples were cross-linked with BS ³ and then separated by gradient gel electrophoresis from 5.0 to 7.5% in the presence of sodium dodecyl sulfate, but under non-reducing conditions. Fractions 57 to 68 show a sample volume of 10 μl; fractions from 70 to 77 show a sample volume of 25 μl. Tetrameric recombinant neuraminidase gives bands of about 220 kDa (tetramer) and about 110 kDa (dimer). Dimeric and monomeric recombinant neuraminidase remain visible in the form of bands of about 110 kDa and about 5 kDa, respectively.

 In FIG. Figure 8 shows the result of analysis by polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate of protein samples taken at various stages of the purification process of recombinant neuraminidase. Lane 1 corresponds to marker proteins; lane 2 corresponds to a crude medium (5 μg); lane 3 precipitate in (20-60)% sodium sulfate (5 μg); lane 4 corresponds to the combined eluate after chromatography on Sepharose Q (2.5 μg); lane 5 corresponds to the combined eluate after separation on N- (p-aminophenyl) oxamic acid — agarose (1 μg); lane 6 corresponds to the combined eluate of tetrameric and dimeric recombinant neuraminidase fractions after gel filtration on Superdex 200 (1 μg).

 In FIG. Figure 9 shows a comparative analysis of the content of carbohydrate associated with natural pronase-treated neuraminidase and carbohydrate associated with recombinant neuraminidase, which is carried out on the basis of fermentation of neuraminidase-glucanase and analysis by polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate. The neuraminidase-glucanase enzyme is visible as a band of about 35 kDa. In FIG. 9A lane 1 corresponds to marker proteins; lane 2 — non-fermented natural neuraminidase (1 μg); lane 3 - natural neuraminidase treated with neuraminidase-glucanase (1 μg). In FIG. 9B lanes 1 and 8 correspond to marker proteins; lane 2 — non-fermented tetrameric recombinant neuraminidase; lane 3 — tetrameric recombinant neuraminidase treated with neuraminidase-glucanase; lane 4 — non-fermented dimeric recombinant neuraminidase; lane 5 - dimeric recombinant neuraminidase treated with neuraminidase-glucanase; lane 6 — unfermented monomeric recombinant neuraminidase; lane 7 - monomeric recombinant neuraminidase treated with neuraminidase-glucanase.

; значения для тетрамерной рекомбинантной нейраминидазы обозначены

; значения для димерной рекомбинантной нейраминидазы обозначены

; и значения для мономерной рекомбинантной нейраминидазы обозначены

Антигенный отклик на рекомбинантную нейраминидазу приведен на фиг. 11. Через четырнадцать дней после каждой иммунизации (показаны стрелками) у мышей берут образцы крови и присутствие антител рекомбинантной нейраминидазы определяют методом иммуноферментного твердофазного анализа (детали эксперимента приведены в тексте). Полностью закрашенные прямоугольники и заштрихованные прямоугольники соответствуют средним титрам (± стандартное отклонение) соответственно вакцинированных и контрольных животных. На фиг. 12 показана гомовариантная защита. Вакцинированным [--- в (А);

в (В) и (C)] и контрольным [_ __ в (А); • в (В) и (С)] животным назначают дозу 20 LD₅₀ гомовариантного адаптированного для мышей вируса Х-47. За развитием инфекции следят, регистрируя уровень смертности (А) и измеряя ректальную температуру (В) и вес тела (С) мышей (см. детали эксперимента в тексте). Цифровые значения являются средними (± стандратное отклонение).FIG. 10 shows the antigenicity of recombinant neuraminidase and natural pronase-treated neuraminidase close to identity. Equal concentrations of recombinant neuraminidase and natural neuraminidase proteins are created, and then they are successively diluted in a 1: 2 ratio and enzyme-linked immunosorbent assay is performed. The drawing shows a sinusoidal antigenicity curve obtained for special antigens. Values for natural pronase-treated neuraminidase are indicated by

; values for tetrameric recombinant neuraminidase are indicated

; values for dimeric recombinant neuraminidase are indicated

; and the values for the monomeric recombinant neuraminidase are indicated

The antigenic response to recombinant neuraminidase is shown in FIG. 11. Fourteen days after each immunization (shown by arrows), blood samples were taken from mice and the presence of recombinant neuraminidase antibodies was determined by enzyme-linked immunosorbent assay (the details of the experiment are given in the text). Fully filled rectangles and shaded rectangles correspond to the average titers (± standard deviation) of the vaccinated and control animals, respectively. In FIG. 12 shows homovariant protection. Vaccinated [--- in (A);

in (B) and (C)] and the control [_ __ in (A); • in (B) and (C)] animals are given a dose of 20 LD _{50 of the} homovariant virus adapted to mice X-47. The development of the infection is monitored by recording the mortality rate (A) and measuring the rectal temperature (B) and body weight (C) of the mice (see the text of the experiment for details). Numeric values are average (± standard deviation).

в (В) и (С)] и контрольным [___ в (А); • в (В) и (С)] животным назначают дозу 20 LD₅₀ гетеровариантного адаптированного для мышей вируса Х-31. За развитием инфекции следят, регистрируя уровень смертности (А) и измеряя ректальную температуру (В) и вес тела (С) мышей (см. детали эксперимента в тексте). Цифровые значения являются средними (± стандартное отклонение).A heterovariant protection is shown in FIG. 13. Vaccinated [--- in (A);

in (B) and (C)] and control [___ in (A); • in (B) and (C)] animals are prescribed a dose of 20 LD ₅₀ heterovariant virus-adapted for mice X-31. The development of the infection is monitored by recording the mortality rate (A) and measuring the rectal temperature (B) and body weight (C) of the mice (see the text of the experiment for details). Numeric values are average (± standard deviation).

в (В) и (C)] и контрольной сыворотки [__ _ в (А); • в (В) и (С)]. Через двадцать четыре часа мышам вводят дозу 20 LD₅₀ адаптированного для мышей вируса Х-47 (см. детали эксперимента в тексте). Показаны соответственно уровень смертности (А), ректальная температура (В) и вес тела (С) мышей. Цифровые значения являются средними (± стандартное отклонение).FIG. 14 illustrates passive immunization. Groups of mice are passively immunized with an intraperitoneal injection of immune serum recombinant neuraminidase [--- in (A);

in (B) and (C)] and control serum [__ _ in (A); • in (B) and (C)]. Twenty-four hours later, a dose of 20 LD ₅₀ of the X-47 virus adapted for mice was administered to mice (see text for experimental details). Mortality (A), rectal temperature (B) and body weight (C) of mice are shown respectively. Numeric values are average (± standard deviation).

 In FIG. 15 is a diagram of plasmid pP1C9, which, in addition to the AOX1 promoter sequence and terminator, contains the H1S4 marker from P. pastoris and the signal for secretion of the alpha factor prepro gene from Saccharomyces cerevisiae. Behind the secretion signal is a compound cloning site.

In FIG. Figure 16 gives an overview of the fusion modes between the prepro-signal of secretion and the recombinant part of the “cap” of neuraminidase. "KEX2" indicates the place where the propeptide is cleaved according to the Golgi using endogenous KEX-2 protease. The dipeptide (Glu-Ala) ₂ is removed using a STE13 type dipeptidyl aminopeptidase. The tyrosine residue is not introduced from neuraminidase, but is not removed. The next proline corresponds to position 79 of neuraminidase from X-47.

 In FIG. Figure 17 shows a Western blot analysis on a 12.5% polyacrylamide gel of five samples of medium of individual transformed cells. Lane 1 contains a medium sample of an untransformed P. pastoris strain. A substance isolated from 1 ml of culture medium by precipitation with trichloroacetic acid is placed on each lane.

 Tab. 1 refers to one typical cleaning experiment (see text for details). The volume after Superdex 200 corresponds to the combined amount after two chromatography operations.

 Tab. 2 refers to one typical cleaning experiment (see test for details). The indicated volumes after gel filtration on Superdex 200 correspond to the combined fractions of recombinant neuraminidase after two chromatographic operations.

Claims (5)

 1. Recombinant NA2 type influenza neuraminidase obtained by culturing in a suitable culture medium insect cells infected with a virus transformed by homologous recombination and its genome using the pAC21VNAS expression vector (LMBP2976).  2. Recombinant influenza neuraminidase according to claim 1, characterized in that the insect cells are insect sf9 cells.  3. Recombinant influenza neuraminidase according to claim 1 or 2 for use in an influenza vaccine.  4. Recombinant NA 2 type influenza neuraminidase obtained by culturing in a suitable culture medium of yeast cells transformed with the IVNAfls pPP1 expression vector (LMBP3223).  5. Recombinant influenza neuraminidase according to claim 4 for use in an influenza vaccine.

Images