Influenza T-cell immunization against diverse influenza A viruses

Abstract

There is provided a T cell-based universal influenza vaccine including internal genes capable of preparing against infection by broad hetero-subtypic influenza viruses, and thereby preparing for unpredictable epidemic influenza. The present invention selected internal genes of the consensus sequence obtained from bird, pig, and human influenza isolates in order to develop the T cell-based universal flu vaccine. The T cell-based universal flu vaccine according to the present invention is characterized by including at least one CTL epitope, by containing a plurality of internal genes and using the entire sequence of the internal gene itself. The T cell-based universal flu vaccine can achieve broad defense in infection with hetero-subtypic influenza viruses.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS

This application claims priority under 35 U.S.C. §119 to Korean Patent Application No. 10-2012-0033317, filed on Mar. 30, 2012, in the Korean Intellectual Property Office, the disclosure of which is incorporated herein by reference in its entirety.

TECHNICAL FIELD

The following disclosure relates to a T cell-based universal influenza vaccine capable of preparing against infection by broad hetero-subtypic influenza viruses and thereby preparing for unpredictable epidemic influenza.

BACKGROUND

Since the 20^th century, influenza infectious diseases have claimed a huge number of victims (Nat. Immunol 7,449-455, 2006). Although it has been possible to predict variant viruses that will prevail in the season next year due to the preparation for seasonal influenza by the Global Influenza Surveillance Network (GISN), there still remains uncertainty in that the predicted vaccine strains do not correspond to the epidemic strains of the following year or recombinant vaccine strains are impossible to mass-produce. The epidemic of the Fujian strain from 2003 to 2004, a variant of vaccine strain A/Panama/2007/99, caused the annual average infant mortality rate to rapidly increase from 9 to 153 in the following year (N Engl J Med 350,218-220, 2004). This may be a representative example in which the predicted vaccine candidate strain does not match to the epidemic strain. Unfortunately, the known existing egg-based production process failed to overcome the above disadvantages in the production of influenza vaccine.

The Spanish flu notoriously increased the mortality rate due to unprecedented extreme toxicity. In addition, appearance of highly pathogenic avian influenza (HPAI) H5N1, which has similar pathogenicity to the Spanish flu highlighted the fact that it is very difficult and impossible to prepare for infectious diseases that will prevail in the future (Nature 437, 889-893, 2005). The Global Influenza Surveillance Network (GISN) has conducted epidemiological survey of H5N1 since 2004, and regularly updated the vaccine candidates. Most vaccine manufacturers have developed H5N1 prepandemic vaccine according to the prediction of GISN, and currently, this is the only measure.

The pandemic H1N1 and highly pathogenic avian influenza (HPAI) H5N1 in 2009 caused concerns that highly pathogenic and highly infectious recombinant influenza reassortants might appear in the future. Eventually, only the universal influenza vaccine capable of preventing infection by broad types of viruses would be the ultimate solution. There was a study of proposing a M2 protein as a promising target for universal influenza vaccine (Nat Med 5, 1157-1163, 1999). Since the M2 protein exhibits a significant level of expression on a surface of the cell infected with virus, it can prevent disease or death due to highly pathogenic influenza viruses by using anti-M2 antibodies in cell mediated immunity for removing the cells infected with viruses, but there are some limits therein.

In addition, the combinations of adjuvant agents, virus-like particles (VLP), or inactive seasonal influenza vaccine, for enhancing immunogenicity of M2, are being investigated.

Recently, a highly conserved hemagglutinin (HA) stem region of the influenza virus is drawing attention, and a relatively conserved HA stem is receiving attention as an attractive target capable of neutralizing viruses broadly. Studies about new vaccines against different types of influenza by using these among several groups are being conducted. However, anti-HA stem antibodies have relatively low neutralization activity similarly to the anti-M2 antibodies because they are not direct neutralizing antibodies. Therefore, as long as there are not other defense cooperation means in an antibody-meditated cell immunity aiming on defense, the anti-HA stem antibodies are effective only when the challenge dose is low. Various approaches for T cell-based vaccine have been attempted even though there were no commercially successful cases. Recent studies about HIV and HCV vaccines represented several promising antigen delivery means for inducing broad and active T cell immunity at the time of defense against highly variable viruses. According to the results of study about historical events of the Asian influenza, which was conducted in Cleveland in the year 1957, it can be found that the presence of cross-reactive memory T cells induced by the existing infected viruses has a defense effect even when hetero-subtypic influenza newly prevail. T-cell response has been supposed as a cross defense immune reaction against influenza infection in many clinical researches and animal studies including ferrets and monkeys. The increase in T-cell immunity due to re-infection by hetero-subtypic influenza viruses eventually showed that, the higher the T-cell immunity, the more efficient the defense against highly pathogenic influenza infection. Many research groups have studied antigen delivery methods by highly conserved internal proteins or epitope of influenza virus, such as NP or M1, by using DNA, adenovirus, poxvirus, or live attenuated virus.

Epstein researchers showed that recombinant adenovirus expressing NP and M2 had surprising efficacy in the defense against hetero-subtypic viruses when intranasal administration of vaccine was conducted and a fatal dose of challenge was attempted. T-cell vaccine has an advantage of fast induction for defense immunity, and in particular, the defense immunity is induced in one week after immunization. Many studies have proved that vaccinia virus has promising possibility as a T-cell vaccine vector.

PRIOR ART DOCUMENTS

Non-Patent Documents

• 1. (Nat Immunol 7,449-455, 2006)
• 2. (N Engl J Med 350,218-220, 2004)
• 3. (Nature 437,889-893, 2005)
• 4. (Nat Med 5, 1157-1163, 1999)

SUMMARY

An embodiment of the present invention is directed to providing a T-cell based universal influenza vaccine containing internal genes allowing broad defense against various hetero-subtypic variants of influenza viruses.

The present invention provides a T-cell based universal influenza vaccine containing internal genes capable of defending broad types of influenza variants in order to prepare for unpredictable epidemic influenza.

Further, the present invention provides a T-cell based universal influenza vaccine containing, as effective components, five kinds of internal genes, NP (SEQ ID NO: 1), PA (SEQ ID NO: 3), PB1 (SEQ ID NO: 5), PB2 (SEQ ID NO: 7), and M1 (SEQ ID NO: 9), which have a consensus sequence and are selected from the bird, pig, and human influenza isolates. An aspect of the present invention provides a T-cell based universal influenza vaccine containing, as effective components, NP (SEQ ID NO: 1) and PA (SEQ ID NO: 3) genes. Another aspect of the present invention provides a T-cell based universal influenza vaccine containing, as effective components, NP (SEQ ID NO: 1) and PA (SEQ ID NO: 3) genes, and further containing, as another effective component, at least one selected from PB1 (SEQ ID NO: 5), PB2 (SEQ ID NO: 7), and M1 (SEQ ID NO: 9) genes. Still another aspect of the present invention provides a T-cell based universal influenza vaccine containing all of NP (SEQ ID NO: 1), PA (SEQ ID NO: 3), PB1 (SEQ ID NO: 5), PB2 (SEQ ID NO: 7), and M1 (SEQ ID NO: 9) genes.

In addition, in the T-cell based universal influenza vaccine:

a CTL epitope of the NP gene has a sequence consisting of:

(SEQ ID NO: 11)
1. MASQGTKRSYEQMET,
(SEQ ID NO: 12)
2. GIGRFYIQMCTELKL,
(SEQ ID NO: 13)
3. MVLSAFDERRN,
(SEQ ID NO: 14)
4. YLEEHPSAGKDPKKTGGPIY,
(SEQ ID NO: 15)
5. LYDKEEIRRIWRQANNG,
(SEQ ID NO: 16)
6. ATYQRTRAL,
(SEQ ID NO: 17)
7. YERMCNILKG,
(SEQ ID NO: 18)
8. QVRESRNPGNAEIEDLIFLA,
(SEQ ID NO: 19)
9. QLVWMACHSAAFEDLRVSSF,
or
(SEQ ID NO: 20)
10. QPTFSVQRNLPF;

a CTL epitope of the PA gene has a sequence consisting of:

(SEQ ID NO: 21)
1. KIETNKFAAICTHLEVCFMYSDFHF,
(SEQ ID NO: 22)
2. RTMAWTVVNSI,
(SEQ ID NO: 23)
3. VEKPKFLPDLY,
(SEQ ID NO: 24)
4. YYLEKANKIKSE,
(SEQ ID NO: 25)
5. THIHIFSFTGEEMA,
(SEQ ID NO: 26)
6. RGLWDSFRQSERGEETIEE,
(SEQ ID NO: 27)
7. RSKFLLMDALKLSIE,
(SEQ ID NO: 28)
8. HEGEGIPLYDAIKC,
(SEQ ID NO: 29)
9. SQLKWALGENMA,
(SEQ ID NO: 30)
10. EFNKACELTDSSWI,
(SEQ ID NO: 31)
11. SRPMFLYVRTNGTSK,
or
(SEQ ID NO: 32)
12. AESRKLLLI;

a CTL epitope of the PB1 gene has a sequence consisting of:

(SEQ ID NO: 33)
1. MDVNPTLLFLKVPAQNAISTTFPYTGDPPYSHGTGTGYTMDTVNRTHQYSE,
(SEQ ID NO: 34)
2. MAFLEESHPGIFENS,
(SEQ ID NO: 35)
3. VQQTRVDKLTQGRQTYDWTLNRNQPAATALANTIE,
(SEQ ID NO: 36)
4. TKKMVTQRTIGKKK,
(SEQ ID NO: 37)
5. FVETLARSICEKLEQSGL,
(SEQ ID NO: 38)
6. RMFLAMITYITRNQP,
(SEQ ID NO: 39)
7. LSIAPIMFSNKMARLGKGYMFESKSMKLRTQIPAEMLA,
(SEQ ID NO: 40)
8. SPGMMMGMFNMLSTVLGVS,
(SEQ ID NO: 41)
9. GINMSKKKSYIN,
(SEQ ID NO: 42)
10. TGTFEFTSFFYRYGFVANFSMELPSFGVSGINESADMSI,
(SEQ ID NO: 43)
11. GVTVIKNNMINNDLGPATAQMALQLFIKDYRYTYRCHRGDTQIQTRRSFE,
(SEQ ID NO: 44)
12. VSDGGPNLY,
(SEQ ID NO: 45)
13. MEYDAVATTHSW,
(SEQ ID NO: 46)
14. PKRNRSILNTSQRGILEDEQMYQ,
or
(SEQ ID NO: 47)
15. AEIMKICST;

a CTL epitope of the PB2 gene has a sequence consisting of:

(SEQ ID NO: 48)
1. LMSQSRTREILTKTTVDHMAIIKKYTSGRQEKNP,
(SEQ ID NO: 49)
2. WMMAMKYPI,
(SEQ ID NO: 50)
3. PERNEQGQTLWSK,
(SEQ ID NO: 51)
4. PLAVTWWNRNGP,
(SEQ ID NO: 52)
5. GPVHFRNQVKIRR,
(SEQ ID NO: 53)
6. YIEVLHLTQGTCW,
(SEQ ID NO: 54)
7. EQMYTPGGEV,
(SEQ ID NO: 55)
8. NDDVDQSLIIAARNIVRRA,
(SEQ ID NO: 56)
9. ASLLEMCHSTQIGG,
(SEQ ID NO: 57)
10. SFSFGGFTFK,
(SEQ ID NO: 58)
11. LTGNLQTLK,
(SEQ ID NO: 59)
12. RVHEGYEEFTMVG,
(SEQ ID NO: 60)
13. RATAILRKATRR,
(SEQ ID NO: 61)
14. VAMVFSQEDCM,
(SEQ ID NO: 62)
15. KAVRGDLNF,
(SEQ ID NO: 63)
16. VNRANQRLNPMHQLLRHFQKDAKVLF,
(SEQ ID NO: 64)
17. RVSKMGVDEYS,
(SEQ ID NO: 65)
18. GNVLLSPEEVSETQG,
(SEQ ID NO: 66)
19. LTITYSSSMMWEINGPESVL,
(SEQ ID NO: 67)
20. NTYQWIIRNWE,
(SEQ ID NO: 68)
21. MLYNKMEFEPFQSLVPKA,
(SEQ ID NO: 69)
22. LGTFDTVQIIKLLPFAAAPP,
(SEQ ID NO: 70)
23. QSRMQFSSLTVNVRGSGMRILVRGNSPVFNYN,
or
(SEQ ID NO: 71)
24. GVESAVLRGFLI;
and

a CTL epitope of the M1 gene has a sequence consisting of:

(SEQ ID NO: 72)
1. SLLTEVETYVL,
(SEQ ID NO: 73)
2. KTRPILSPLTKGIL,
(SEQ ID NO: 74)
3. GFVFTLTVPSE,
(SEQ ID NO: 75)
4. LYRKLKREITF,
(SEQ ID NO: 76)
5. ALASCMGLIY,
(SEQ ID NO: 77)
6. MGTVTTEVAFGLVCA,
(SEQ ID NO: 78)
7. NRMVLASTTAKAMEQMAGSS,
or
(SEQ ID NO: 79)
8. QARQMVQAMR.

Therefore, the NP (SEQ ID NO: 1), PA (SEQ ID NO: 3), PB1 (SEQ ID NO: 5), PB2 (SEQ ID NO: 7), and M1 (SEQ ID NO: 9) genes each are characterized by containing at least one CTL epitope.

Preferably, a T-cell vaccine delivery vector may include a recombinant virus, and the recombinant virus is selected from recombinant vaccinia virus, recombinant adenovirus, recombinant adeno associated virus, recombinant retrovirus, and recombinant lentivirus. More preferably, the recombinant (replicating) vaccinia virus (hereinafter, referred to as ‘rVV’) may be used.

The administration route of the vaccine according to the present invention may be a skin scarification (s.s.), intramuscular (i.m.), intranasal (i.n.), intradermal (i.d.), intravenous (i.v.), or intraperitoneal (i.p.) route, but is not limited thereto. An administration method using the skin scarification (s.s.) route is most preferable.

The T-cell based universal influenza vaccine of the present invention may include at least one additive selected from pharmaceutically acceptable immunopotentiator, carrier, excipient, and diluent, and may be provided in a unit-dose container or a multi-dose container, for example, sealed ample, bottle, or the like.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 shows graph depicting T-cell immunogenicity according to the concentration of rVV-NP vaccine administered to C57BL/6 mice by skin scarification. rVV-EGFP as a negative control was administered by skin scarification, and x-31 as a positive control was intranasally administered. The splenocytes stimulated with NP₃₆₆ CTL epitope peptides after 4 weeks of immunization were analyzed by IFN-γ-ELISPOT assay (each bar indicates average±standard deviation (SD) for 2 mice/group.

FIG. 2 shows a graph depicting the effect according to the administration method of rVV-NP immunization by using IFN-γ-ELISPOT activity.

10⁶ pfu of rVV-NP vaccine was administered to the C57BL/6 mice through the intramuscular (i.m), intranasal (i.n.), or skin scarification (s.s.) route. The splenocytes stimulated with NP₃₆₆ CTL epitope peptides after 10 days or 4 weeks of immunization were analyzed by IFN-γ-ELISPOT assay (each bar indicates average±standard deviation (SD) for 2 mice/group.

FIG. 3 shows a graph depicting the effect according to the vaccination administration method of x-31 by using IFN-γ-ELISPOT activity. C57BL/6 mice were vaccinated with 10⁴ pfu of x-31 through the combination of intramuscular (i.m) and intranasal (i.n.) routes two times at an interval of 4 weeks, without anesthesia. The splenocytes stimulated with by NP₃₆₆ CTL epitope peptides after 10 days of vaccination were analyzed by IFN-γ-ELISPOT assay (each bar indicates average±standard deviation (SD) for 2 mice/group.

FIG. 4A shows a graph depicting the influenza-specific T-cell response induced by rVV-flu vaccination route in x-31-prime-immunized mice. x-31-primed C57BL/6 mice were challenged with 10⁶ pfu of vaccinia influenza. rVV-flu (5) is vaccine consisting of rVV-NP, -PA, -M1, -PB2, and -PB1, and parental vaccinia (VV) or x-31 is a control. FIG. 4A shows evaluation of IFN-γ-ELISPOT activity in splenocytes stimulated with NP₃₆₆ or PA₂₂₄ epitope or PR8-infected EL4 cells after 3 weeks of immunization in C57BL/6 mice.

FIG. 4B shows a graph depicting the influenza-specific T-cell response induced by rVV-flu vaccination route in x-31-prime-immunized mice. x-31-primed C57BL/6 mice were challenged with 10⁶ pfu of vaccinia influenza. rVV-flu (5) is vaccine consisting of rVV-NP, -PA, -M1, -PB2, and -PB1, and parental vaccinia (VV) or x-31 is a control. FIG. 4B shows evaluation of IFN-γ-ELISPOT activity in lung cells stimulated with NP₃₆₆ or PA₂₂₄ epitope or PR8-infected EL4 cells after 3 weeks of immunization in C57BL/6 mice.

FIG. 4C shows a graph depicting the influenza-specific T-cell response induced by rVV-flu vaccination route in x-31-prime-immunized mice. x-31-primed C57BL/6 mice were challenged with 10⁶ pfu of vaccinia influenza. rVV-flu (5) is vaccine consisting of rVV-NP, -PA, -M1, -PB2, and -PB1, and parental vaccinia (VV) or x-31 is a control. FIG. 4C shows functional lytic activity by in vivo CTL assay, on day 7 after immunization (each bar indicates average±standard deviation (SD) for 2˜3 mice/group).

FIG. 4D shows a graph depicting the influenza-specific T-cell response induced by rVV-flu vaccination route in x-31-prime-immunized mice. Balb/c mice were challenged with 10⁶ pfu of vaccinia influenza. rVV-flu (5) is vaccine consisting of rVV-NP, -PA, -M1, -PB2, and -PB1, and parental vaccinia (VV) or x-31 is a control. FIG. 4D shows evaluation o_{f I}FN-γ-E_LISPOT activity in splenocytes stimulated with NP₁₄₇ epitope after 3 weeks of immunization in Balb/c mice.

FIG. 4E shows a graph depicting the influenza-specific T-cell response induced by rVV-flu vaccination route in x-31-prime-immunized mice. Balb/c mice were challenged with 10⁶ pfu of vaccinia influenza. rVV-flu (5) is vaccine consisting of rVV-NP, -PA, -M1, -PB2, and -PB1, and parental vaccinia (VV) or x-31 is a control. FIG. 4E shows ev^aluation of IFN-γ-ELISPOT activity in lung cells stimulated with NP₁₄₇ epitope after 3 weeks of immunization in Balb/c mice.

FIG. 4F shows a graph depicting the influenza-specific T-cell response induced by rVV-flu vaccination route in x-31-prime-immunized mice. Balb/c mice were challenged with 10⁶ pfu of vaccinia influenza. rVV-flu (5) is vaccine consisting of rVV-NP, -PA, -M1, -PB2, and -PB1, and parental vaccinia (VV) or x-31 is a control. FIG. 4F shows functional lytic activity by in vivo CTL assay, on day 7 after immunization (each bar indicates average±standard deviation (SD) for 2˜3 mice/group).

FIG. 5A shows a graph depicting secretion of multifunctional cytoki_ne from influenza-specific T-cells. The splenocytes of C57BL/6 mice were used, and stimulated with NP₃₆₆ (upper panel) or PA₂₂₄ (lower panel) epitopes after 7 weeks of rVV-flu immunization. Secretion of IFN-γ, TNF-α and IL-2 cytokines and expression of degranulation marker (CD107a) were measured by in vivo cytokine staining: FIG. 5A shows the rate (%) of CD8 T-cells secreted by each cytokine.

FIG. 5B shows a graph depicting secretion of multifunctional cytoki_ne from influenza-specific T-cells. The splenocytes of C57BL/6 mice were used, and stimulated with NP₃₆₆ (upper panel) or PA₂₂₄ (lower panel) epitopes after 7 weeks of rVV-flu immunization. Secretion of IFN-γ, TNF-α and IL-2 cytokines and expression of degranulation marker (CD107a) were measured by in vivo cytokine staining: FIG. 5B shows measurement results of T-cell response and degranulation for various cytokine combina_tions in order to anal_yze multi-functionality of the influenza-specific T-cells (each bar indicates average±standard deviation (SD) for 2 mice/group).

FIG. 5C shows a graph depicting secretion of multifunctional cytokine from influenza-specific T-cells. The splenocytes of C57BL/6 mice were used, and stimulated with NP₃₆₆ (upper panel) or PA₂₂₄ (lower panel) epitopes after 7 weeks of rVV-flu immunization. Secretion of IFN-γ, TNF-α and IL-2 cytokines and expression of degranulation marker (CD107a) were measured by in vivo cytokine staining: FIG. 5C shows dot plots for a positive (+) control and rVV-NP+PA.

FIG. 6 shows a graph depicting weight loss for 7 days when C57BL/6 mice were challenged with PR8 (0.5/1_0/10_{0 L}D₅₀), after immunization (each bar indicates average±standard error of measurement (SEM) for 4 mice/group).

FIG. 7A shows graph depicting weight loss and survival rate when x-31-prime-immunized C57BL/6 mice were challenged with PR8 (100 LD₅₀), after 3 weeks of rVV-flu immunization. The % weight loss were tracked for 2 weeks (each bar indicates average±standard error of measurement (SEM) for 8 mice/group.

FIG. 7B shows graph depicting weight loss and survival rate when x-31-prime-immunized Balb/mice were challenged with PR8 (100 LD₅₀), after 3 weeks of rVV-flu immunization. The % weight loss were tracked for 2 weeks (each bar indicates average±standard error of measurement (SEM) for 8 mice/group).

FIG. 7C shows a graph depicting weight loss and survival rate when x-31-prime-immunized C57BL/6 mice were challenged with PR8 (100 LD₅₀), after 3 weeks of rVV-flu immunization. The % survival were tracked for 2 weeks (each bar indicates average±standard error of measurement (SEM) for 8 mice/group.

FIG. 7D shows a graph depicting weight loss and survival rate when x-31-prime-immunized Balb/c mice were challenged with PR8 (100 LD₅₀), after 3 weeks of rVV-flu immunization. The % survival were tracked for 2 weeks (each bar indicates average±standard error of measurement (SEM) for 8 mice/group).

FIG. 7E shows a graph depicting weight loss and survival rate when x-31-prime-immunized C57BL/6 and Balb/c mice were challenged with PR8 (100 LD₅₀), after 3 weeks of rVV-flu immunization. Lung virus titers were measured on day 3, 5, and 7 post challenge (each bar indicates average±standard error of measurement (SEM) for 3 mice/group).

FIG. 7F shows a graph depicting weight loss and survival rate when x-31-prime-immunized C57BL/6 and Balb/c mice were challenged with PR8 (100 LD₅₀), after 3 weeks of rVV-flu immunization. Lung virus titers were measured on day 3, 5, and 7 post challenge (each bar indicates average±standard error of measurement (SEM) for 3 mice/group).

FIG. 8A shows a graph depicting the weight loss and survival rate of x-31-prime-immunized C57BL/6 mice, which were challenged with HPAI H5N1 after 3 weeks of rVV-flu immunization (100 LD₅₀ A/MD/W401/11 (HPAI H5N1 field isolate)). The % weight loss were tracked for 2 weeks (each bar indicates average±standard error of measurement (SEM) for 8 mice/group).

FIG. 8B shows graph depicting the weight loss and survival rate of x-31-prime-immunized C57BL/6 mice, which were challenged with HPAI H5N1 after 3 weeks of rVV-flu immunization (100 LD₅₀ A/MD/W401/11 (HPAI H5N1 field isolate)). The % survival were tracked for 2 weeks (each bar indicates average±standard error of measurement (SEM) for 8 mice/group).

FIG. 8C shows a graph depicting the weight loss and survival rate of x-31-prime-immunized C57BL/6 mice, which were challenged with HPAI H5N1 after 3 weeks of rVV-flu immunization (100 LD₅₀ A/MD/W401/11 (HPAI H5N1 field isolate)). Lung virus titers were measured on day 3, 5, and 7 (each bar represents average±standard error of measurement (SEM) for 3 mice/group).

FIG. 9A shows a graph depicting cross-reactive IFN-γ-ELISPOT activity corresponding to CTL epitope, after the prime-immunized Phil/2/82 mice were subjected to rVV-flu immunization. The cross-reactive IFN-γ-ELISPOT activity was measured on CTL epitope of the priming virus (Phil/2/82) and the challenging virus (Cal/04/09). The above assay was conducted after 4 weeks of priming. (each bar indicates average±standard deviation (SD) for 2 mice/group).

FIG. 9B shows a graph depicting cross-reactive IFN-γ-ELISPOT activity corresponding to CTL epitope, after the prime-immunized Phil/2/82 mice were subjected to rVV-flu immunization. The cross-reactive IFN-γ-ELISPOT activity was measured on CTL epitope of the priming virus (Phil/2/82) and the challenging virus (Cal/04/09). The above assay was conducted after 3 weeks of rVV-flu vaccination (each bar indicates average±standard deviation (SD) for 2 mice/group).

FIG. 10A shows a graph depicting the weight loss and survival rate of Phil/2/82-prime-immunized C57BL/6 mice, which were challenged with pandemic H1N1 adapted into mice after rVV-flu immunization. The mice were challenged with Cal/04/09 virus adapted into 100 LD₅₀ mice, after 3 weeks of immunization, and the % weight loss were observed (each bar indicates average±standard error of measurement (SEM) for 8 mice/group).

FIG. 10B shows a graph depicting the weight loss and survival rate of Phil/2/82-prime-immunized C57BL/6 mice, which were challenged with pandemic H1N1 adapted into mice after rVV-flu immunization. The mice were challenged with Cal/04/09 virus adapted into 100 LD₅₀ mice, after 3 weeks of immunization, and the % survival thereof were observed (each bar indicates average±standard error of measurement (SEM) for 8 mice/group).

DETAILED DESCRIPTION OF EMBODIMENTS

Hereinafter, the present invention will be described in detail with reference to examples and drawings. The examples are only for specifically explaining the present invention and it is obvious to those skilled in the art that the scope of the present invention is not limited by the examples according the gist of the present invention.

Preparation of Animal Model and Virus

C57BL/6 and Balb/c female mice of 6 weeks of age purchased from the Charles-River Laboratory (Orient Bio Inc., Sungnam, Korea) were used in the present invention. Parental NYCBH vaccinia variants were used for production of rVV-flu, and A/PR/8/34 (H1N1) and x-31 (H3N2) influenza viruses were used in the present invention. All procedures for using the H5N1 virus were conducted in a Biosafety Level-3 (BSL-3) facility.

Selection of Internal Genes Containing Consensus Sequence and Representative Sequence

In order to determine the equidistant consensus sequence from bird, pig, and human influenza isolates, a total of 14,011 sequences of the influenza internal genes (PB2, PB1, PA, NP, M1, M2, NS1 and NS2) were searched from influenza virus data of the U.S. National Center for Biotechnology Information (NCBI) (http://www.ncbi.nlm.nih.gov/genomes/FLU/FLU.html).

In respective species, the individual gene phylogenetic tree was created by using a neighbor-joining method, and, finally, the consensus sequence of eight kinds of internal genes was determined. Three among the eight kinds of internal genes, NS1, NS2, and M2, were excluded in selection of the internal genes for the T-cell based universal flu vaccine of the present invention due to the following reasons:

• (1) Since three kinds of internal genes are shorter than the other internal genes, the opportunities for presenting a defense T-cell epitope are limited;
• (2) Since NS1 has been known to suppress Type I interferon response, it may influence the immunogenicity formation when a delivery vector using NS1 as an antigen is used; and
• (3) M2 does not contribute to the defense against a fatal dose of vaccination since the efficacy caused by the combination of rVV-NP+PA and rVV-flu (5) is significantly lower than the efficacy induced by single r-VV-M2 vaccination in an anti-M2 antibody response.

Therefore, five kinds of internal genes, NP, PA, PB1, PB2, and M1 were selected for the T-cell based universal flu vaccine of the present invention, and immunogenicity and efficacy thereof were analyzed.

GenBank Accession Numbers of the five kinds of internal genes and proteins translated therefrom are as follows:

• NP gene (SEQ ID NO: 1), gi:323984, gb:M22574.1,
• NP protein (SEQ ID NO: 2), gi:323985, gb:AAA43095.1;
• PA gene (SEQ ID NO: 3), gi:78097605, gb:CY005610.1,
• PA protein (SEQ ID NO: 4), gi:78097606, gb:ABB20301.1;
• PB1 gene (SEQ ID NO: 5), gi:82654856, gb:CY005794.1,
• PB1 protein (SEQ ID NO: 6), gi:82654857, gb:ABB88367.1;
• PB2 gene (SEQ ID NO: 7), gi:134044357, gb:CY011035.2,
• PB2 protein (SEQ ID NO: 8), gi:113901268, gb:ABI47980.1; and
• M1 gene (SEQ ID NO: 9), gi:4584881, gb:AAD25173.1,
• M1 protein (SEQ ID NO: 10), gi:4584882, gb:AAD25173.1,

TABLE 1
Numbers of influenza isolates and subgroups for determining
final consensus sequences of influenza A internal genes.
	Number of subgroups	Final consensus
Number of influenza	(selection criteria)	sequence
isolates analyzed	(% Ave.a.a.identy ± SD)	Average
	Human	Pig	Bird	Total	Human	Pig	Bird	Total	Distance
PB2	866	187	1303	2356	32 (>97%)	13	36	81	97.70%
						−96%	−97%
PB1	835	199	1213	2247	30 (>97%)	7 (>91%)	45 (>98%)	82	98.70%
PA	826	190	1403	2419	18 (>97%)	12 (>94%)	31 (>96%)	61	9790%
NP	581	187	920	1688	29 (>97%)	27 (>96%)	55 (>96%)	111	96.80%
M1	255	130	525	910	19 (>97%)	18 (>97%)	28	65	97.70%
							−96%
M2	372	137	536	1045	32 (>91%)	32 (>93%)	29 (>91%)	93	92.50%
					(95.0% ± 2.4%)	(95.5% ± 1.5%)	94.7% ± 1.8%)
NS1	710	228	1419	2357	30 (>94%)	32 (>89%)	76 (>90%)	138	90.80%
						(95.8% ± 2.6%)	(96.5% ± 2.2%)
NS2	256	111	622	989	20 (>97%)	15(>91%)	19(>91%)	54	94.40%
Total	14,011		685	97.10%

The T-cell based universal flu vaccine containing the five kinds of internal genes has higher immunogenicity than the control group. When comparing the conventionally known conserved human T-cell epitope with the consensus sequence from the influenza isolates, 1,142 amino acids among 1,147 amino acids of the conserved human T-cell epitope are 99.6% identical to the consensus sequence.

TABLE 2
Homology between known common epitope and
consensus sequence of T-cell based universal flu vaccine
of the present invention
	Total number of amino	Number of amino acids of
	acids of epitope	homologous epitope	Homology %
PB2	366	365	99.7%
PB1	359	358	99.7%
PA	171	170	99.4%
NP	149	148	99.3%
M1	102	101	99.0%
Total	1147.	1142	99.6%

The internal genes of the present invention have average 95% sequence homology to the new influenza (2009 H1N1) (95.3% to A/Mexico/InDRE4114/2009; 95.2% to A/Canada-AB/RV1532/2009; and 95.0% to A/New York/1669/2009), and average 97.8% amino acid homology to the influenza isolates from influenza virus data of the U.S. National Center for Biotechnology Information (NCBI). The above homology level means that the present invention can achieve broad T-cell-mediated defense against some unknown epidemic influenza.

TABLE 3
Homology between influenza isolates from
influenza virus data of the U.S. National Center for
Biotechnology Information (NCBI) and consensus sequence
of T-cell based universal flu vaccine of the present invention
			% Homology of influenza variant
			having consensus sequence (% range of
	Number		influenza isolates in influenza database)*
	of				Mallard/	Philippinne/	California/
	amino	100%	Average	PR/8/34	W401/11	2/82	04/09
	acids	identical	distance	(H1N1)	(H5N1)	(H3N2)	(H1N1)
PB2	759	A/blue-winged	97.7%	97.2%	98.4%	95.5%	97.8%
		Teal/Ohio/926/		(63.5%)	(57.8%)	(75.6%)	(59.3%)
		2002 (H3N8)
PB1	758	A/turkey/Italy/4169/	98.7%	97.6%	98.5%	98.8%	97.1%
		1999 (H7N1)		(87.5%)	(63.9%)	(63.3%)	(87.9%)
PA	716	A/chicken/	97.9%	97.2%	99.2%	95.4%	97.5%
		Hong Kong/		(65.2%)	(38.1%)	(81.2%)	(63.5%)
		17/1977 (H6N1)
NP	498	A/duck/Bavaria/2/	96.8%	94.6%	98.0%	92.4%	94.6%
		1977 (H1N1)		(69.3%)	(55.0%)	(86.4%)	(69.3%)
M1	257	A/chicken/	97.7%	96.8%	96.4%	97.2%	96.5%
		New York/		(67.3%)	(86.0%)	(57.6%)	(67.3%)
		13142-5/94
		(H7N2N5B)
Total	2983	100.0%	97.8%	96.7%	98.1%	95.9%	95.9%
number of				(71.3%)	(56.5%)	(74.1%)	(69.9%)
Amino
acids
*Range (%) of influenza isolates in influenza virus data of the U.S. National Center for Biotechnology Information (NCBI), which share higher homology than variant appeared in the consensus sequence: PB2 (n = 2356), PB1 (n = 2247), PA (n = 2419), NP (n = 1688), M1 (n = 910).

EXAMPLE 1

Production of rVV-flu Vaccine

In order to produce r-VV flu vaccine, influenza gene-optimized human codons were synthesized, and inserted at the EcoRV site of pUc57 (provided by GenScript (Piscataway, N.J.). Each plasmid containing NP or MI gene was cut by AscI/SbfI restriction enzymes, and then the cut pBMSF7C vaccine delivery vector was linked thereto by PstI/AscI. An appropriate size of PB2, PB1, and PA were inserted into the cut fragments of pUC57 by AscI/SbfI, and then the vector was cut by DraI restriction enzyme one more time. The cut pBMSF7C was cloned by PstI/AscI. The rVV-flu vaccine was produced by slightly changing the conventional method thereof (Vaccine 25,630-637, 2007). BHK-21 cells infected with vaccinia virus at multiplicitiy of infection (MOI) of 10 were transfected with pBMSF7C-flu. The viruses were harvested after 24 hours of transfection, and RK-13 cells, from which recombinant HA negative plaques are to be screened, were again infected therewith. After 3 days of infection, the cells were reacted with turkey RBC at 37° C. for 30 minutes to isolate HA negative plaques. The HA negative plaques were continuously cloned until HA positive plaques disappeared, and then confirmed by HA staining of ˜1,000 plaques per 100-mm dish. The final recombinant viruses were purified, and amplified in BHK-21 cells (0.1 MOI). They were low-temperature stored at −80° C. prior to the use thereof.

It was confirmed whether the influenza genes are inserted into the recombinant vaccinia virus by using the PCR method, and the final clones were confirmed by Western blot and sequencing of binding sites.

EXAMPLE 1-1

Western Blot Assay of rVV-flu

BHK-21 cells were infected with rVV-flu at MOI of 0.1 for 1 hour. Then, the cells were incubated for three days, followed by harvesting. Proteins were extracted with an M-PER buffer solution according to the conventionally known existing method (Pierce, Rockford, Ill.). Thermo-denatured proteins were separated by 4-12% SDS-PAGE (Invitrogen Carlsbad, Calif.), and transferred on the nitrocellulose membrane (Invitrogen Carlsbad, Calif.). The membrane was blocked in a PBS buffer solution containing 0.2% Tween 20 together with 5% nonfat milk for 1 hour, and reacted with primary antibodies.

The following antibodies were used for detection:

• anti-PB2 (Santacruz Carlsbad, Calif.);
• anti-PB1 (Santacruz Carlsbad, Calif.);
• anti-PA (Peptron Inc., Daejeon, Korea);
• anti-NP and anti-M1 (X-31/PR8 immunized mouse serum).

The membrane was incubated together with horseradish peroxidase-labeled anti-mouse IgG antibodies (KPL Gaithersburg, Md.), and stained with ECL plus kit (Amersham Buckinghamshire, U.K.).

EXAMPLE 2

Immunogenicity Evaluation of rVV-flu

EXAMPLE 2-1

IFN-γ ELISPOT Assay

In the present invention, immunogenicity of rVV-flu was evaluated through IFN-γ ELISPOT assay in order to determine the optimum vaccine combination in vaccinia based influenza vaccination. High immunogenicity of nucleoprotein (NP) was confirmed in the H-2b mouse model. Several doses of rVV-NP were administered through the conventionally known skin scarification (s.s.), and immunogenicity due to this was induced according to the administration dose In T-cell response. It was confirmed that the boosting immunization improved the T-cell response (see, FIG. 1).

In the present invention, the administration of vaccine was performed by an intramuscular (i.m.) method, an intranasal (i.n.) method, or a skin scarification (s.s.) method. Here, the skin scarification (s.s.) method was most preferable since it exhibits 5 times higher immunogenicity than the intramuscular (i.m.) method or the intranasal (i.n.) method (see, FIG. 2).

Since 91˜98% of humans have been infected with influenza virus, they have influenza-specific memory T-cells. Therefore, in the present invention, influenza-prime-immunized mice were used as a mouse model for evaluating immunogenicity and efficacy of vaccine. Among four combinations of the intramuscular (i.m.) method and the intranasal (i.n.) method, x-31 intramuscular priming/intranasal boosting administration method induced the highest level of T-cell immunity (see, FIG. 3).

The x-31-prime-immunized C57BL/6 mice were challenged with 10⁶ pfu of vaccinia influenza, and the Balb/c mice were challenged with 10⁶ pfu of vaccinia influenza.

In the C57BL/6 mice, immunogenically predominant epitopes of the following two kinds of T-cell based universal flu vaccine were NP₃₆₆ and PA₂₂₄, and 90% or more of flu-specific T-cell response could be induced (see, FIGS. 4A to 4F).

• rVV-NP+PA
• rVV-flu (5) (Flu (5) means internal genes, NP, PA, PB1, PB2, and M1.)

The epitopes of NP and PA, NP₃₆₆ and PA₂₂₄, exhibited the most predominant immunogenicity in flu-specific T-cell response. However, vaccine containing T-cell epitopes obtained from the five kinds of internal genes is provided, to thereby allow human to prepare for a case where NP and PA are contained as T-cell epitope.

EXAMPLE 3

Immunization Using Mouse Model

5 ul of 10⁴ pfu of X-31 was administered to the mice through the intranasal (i.n.) route without anesthesia (priming immunization, on day 0). The mice were vaccinated with 5 ul of 10⁶ pfu of rVV-flu by skin scarification (s.s.) administration (boosting immunization, on day 28).

The parental-VV was used alone as a negative control. The priming vaccination of a positive control of 50 ul was performed on the thigh muscle at the rear parts of both hind legs through the intramuscular (i.m.) route, and the boosting vaccination was performed by using the intranasal (i.n.) route (10⁴ pfu of x-31).

Eight mice were used for each group. On the day after three weeks of the immunization, the mice were anesthetized with avertin, and infected with 50 ul of A/PR/8/34 virus at LD₅₀ of 0.5/10/100.

As the hetero-subtypic flu virus, 100 LD₅₀ highly pathogenic H5N1 A/Mallard Duck/Korea/W401/11 and epidemic A/California/04/09 were used. The survival and the change in weight of the mice were confirmed everyday for 2 weeks, and the mice were euthanized when they lost 30% or more of the initial weight for animal welfare (see, FIGS. 6, 7A to 7F, 10A and 10B).

EXAMPLE 4

Preparation of Lung Cells for Cell Immunity

The lung cells were disrupted following the conventionally known protocol (Miltenyi Biotec Inc., Carlsbad, Calif.). The disrupted lung cells were treated with RBC lysis buffer on ice for 5 minutes (Sigma-Aldrich). The lung cells with the removal of RBC were disentangled in RPMI, and in vivo apoptosis effect (in vivo CTL) assay and ELISPOT (Enzyme-Linked ImmunoSpot) assay were conducted.

EXAMPLE 5

Ex-Vivo IFN-γ ELISPOT Assay

The ELISPOT assay was conducted using the mouse IFN-γ ELISPOT^PLUS kit (Mabtech, AB Nacta Strand, Sweden). 2×10⁵ fresh splenocytes or lung cells were plated on the 96-well plate coated with anti-mouse IFN-γ capture antibody, and stimulated with 1 μM of influenza peptide epitope (Peptron, Daejeon, Korea) or EL4 (2×10⁵ cells/well) infected with 1 MOI of PR8 virus for 18 hours.

The flu-specific CD8+ T-cell epitope was as follows:

NP366-74    (Db, ASNENMETM);
PA224-33    (Db, SSLENFRAYV);
PB2196-210, (Db or Kb, CKIAPLMVAYMLERE);
PB1703-11   (Kb, SSYRRPVGI);
and
M1128-35    (Kb, MGLIYNRM).

The splenocytes pulsed with PB2_198-206 or infected with B/Malaysia/2506/04 were used in a level of 8±5(SD) and 13±9(SD) per well for a negative control. The lung cells were used in a higher level for a higher negative control (14±13 and 25±26). The number of IFN-γ secretion cells ((ISCs) was measured using an ELISPOT reader (AID, Strassberg, Germany) (See, FIGS. 5A to 5C).

EXAMPLE 6

In Vivo Cytotoxic T-Cell (CTL) Assay

The splenocytes obtained from naive mice were stained with 5 M PKH26 (Sigma-Aldrich, Saint Louis, Mo.) at room temperature for 10 minutes. Labeling was stopped by addition of the same amount of FBS, followed by incubation for 1 minute. Additional staining with 0.5 uM, 2 uM, or 8 uM of CFSE (Carboxy Fluorescein Succinimidyl Ester) was conducted at 37° C. for 10 minutes, and then stopped by addition of the same amount of FBS, followed by incubation for 1 minute. The doubly stained cells were stimulated with the indicator epitope peptide (1 uM) at 37° C. for 1 hour.

Each target cell and a different peptide were mixed together, and the mixture was administered to the naive or immunized mice through intravenous (i.v.) administration. After 12 hours of administration, single cell dispersions of the splenocytes and lung cells were prepared, and activity of in vivo cytotoxic T cell (CTL) was measured using flow cytometry analysis (BD LSRFortessa, Bioscience BD, Piscataway, N.J.). The cell killing ratio (%) was calculated according to the following equation.

$\begin{array}{cc}\mathrm{Cell}\mathrm{killing}\mathrm{ratio}\left(\%\right)=100-\left\{\begin{array}{c}(\mathrm{number}\mathrm{of}\mathrm{stimulated}\mathrm{target}\mathrm{cells}\mathrm{remaining}\\ \mathrm{in}\mathrm{the}\mathrm{immunized}\mathrm{mouse}/\mathrm{number}\mathrm{of}\\ \mathrm{un}\text{-}\mathrm{stimulated}\mathrm{target}\mathrm{cells}\mathrm{remaining}\mathrm{in}\mathrm{the}\\ \mathrm{immunized}\mathrm{mouse})/(\mathrm{number}\mathrm{of}\mathrm{stimulated}\\ \mathrm{target}\mathrm{cells}\mathrm{remaining}\mathrm{in}\mathrm{the}\mathrm{naïve}\mathrm{mouse}/\mathrm{number}\\ \mathrm{of}\mathrm{un}\text{-}\mathrm{stimulated}\mathrm{target}\mathrm{cells}\mathrm{remaining}\mathrm{in}\\ \mathrm{the}\mathrm{naïve}\mathrm{mouse})\times 100\end{array}\right\}& \left[\mathrm{Equation}\right]\end{array}$

EXAMPLE 7

In Vivo Cytokine Staining (ICS) and T Cell Polyfunctionality Analysis

In the present example, polyfunctionality of influenza specific T cell induced by rVV-flu vaccine was observed. After rVV-flue immunization, stimulations with NP₃₆₆(upper panel) and PA₂₂₄ (lower panel) epitopes were conducted for 7 days. Secretion of IFN-γ, TNF-α and IL-2 cytokines and expression of degranulation markers (CD107a) were measured by in vivo cytokine staining (see, FIGS. 5A to 5C).

In the presence of anti-CD107a-PE-Cy7 (BD Biosciences), the splenocytes were stimulated with cytotoxic T cell (CTL) epitope peptide (1 ug/ml) such as NP₃₆₆ or PA₂₂₄, and, after 1 hour, monesin (GolgiStop, BD Biosciences), which is a kind of transferring antibiotic materials for improving cell permeability, was added. After 5 hours of incubation, the splenocytes were stained with ethidium monoazide, and anti-CD3-Horizon V500, anti-CD4-Horizon V450, and anti-CD8-APC-H7 (BD Biosciences) were stained with ethidium monoazide. They were permeabilized using Cytofix/Cytoperm kit (BD Biosciences) and further stained with anti-IL-2-FITC (eBioscience), anti-IFN-γ-APC (BD Biosciences) and anti-TNF-PE (BD Biosciences). FACS assay was conducted using an LSRII flow cytometer (BD Biosciences), and data were analyzed by FlowJo software (Treestar, San Fran Carlos, Calif.). Quantification and analysis of T cells with respect to various combinations of cytokines and degranulation were conducted by using FlowJo software.

Influenza HA is a main target for monitoring antibody-mediated host immunity, and variation thereof continuously occurs. Therefore, only influenza HA having a consensus sequence has an advantage of cross-defense. The internal genes have relatively low immunity than HA, and thus have been significantly conserved. The conserved sequence of the present invention is characterized by having sufficient cell immunity for allowing the cross-reaction since it has average 97.8% homology The present invention is characterized by selecting the entire internal genes in their original states, instead of epitope sequences of providing customized antigen representation of T cell epitope according to individual HLA polymorphism. In addition, the multi-epitope-specific T cell response can decrease the possibility of variant formation. Notwithstanding the use of the multi-gene, the decrease in immunogenicity of individual antigen is due to antigen competition among individual recombinant viruses in replication of T cells and antigen presentation. This was demonstrated from antigen competition in the DNA vaccine combination study. In the present invention, cross-reactive IFN-γ-ELISPOT activity was measured with respect to CTL epitopes of the priming virus (Phil/2/82) and the challenging virus (Cal/04/09). The assay was conducted after 4 weeks of priming (a) and after 3 weeks of rVV-flu vaccination (see, FIGS. 9A and 9B).

EXAMPLE 8

Titer Assay of Lung Virus

The mice infected with influenza were enthanized after 3, 5, and 7 days, to obtain lung tissues thereof. The obtained lung tissues were disrupted in 10% W/V serum free Minimum Essential Me-dia (MEM) for analyzing virus titer using plaque assay. The limit of virus detection was 100 pfu/g of the lung (this value corresponds ˜20 pfu/mouse). The titer of HPAI H5N1 virus was measured by using the eggs in BSL3+ facilities received approval for influenza experiment. The limit of virus detection was 0.7 log 10 EID₅₀/ml or lower (see, FIGS. 7A to 7F, and 8A to 8C).

The differences in weight loss and lung virus titer between groups were analyzed by the Student T test. The difference in survival was calculated by the Fisher's exact test, and it was confirmed that the statically significant p value was 0.05 or lower (see, FIGS. 7A to 7F, and 8A to 8C).

The T cell-based universal influenza vaccine of the present invention includes at least one CTL epitope in each of the internal genes, and thus can enhance immunogenicity thereof according to combination of the internal genes, and thereby can defend against various hetero-subtypic variants of the influenza viruses.

Claims

1. A T cell-based immunogenic composition comprising an influenza M1 gene having the sequence of SEQ ID NO:9, further comprising at least one additive selected from the group consisting of pharmaceutically acceptable immunopotentiator, carrier, excipient, and diluent.

2. The T cell-based immunogenic composition of claim 1, further comprising at least one influenza gene selected from the group consisting of an NP gene having the sequence of SEQ ID NO:1, a PA gene having the sequence of SEQ ID NO:3, a PB1 gene having the sequence described by SEQ ID NO: 5, and a PB2 gene having the sequence described by SEQ ID NO: 7.

3. A T cell-based immunogenic composition comprising all of NP (SEQ ID NO: 1), PA (SEQ ID NO: 3), PB1 (SEQ ID NO: 5), PB2 (SEQ ID NO: 7), and M1 (SEQ ID NO: 9) gene.

4. The T cell-based immunogenic composition of claim 2, wherein the NP (SEQ ID NO: 1), PA (SEQ ID NO: 3), PB1 (SEQ ID NO: 5), PB2 (SEQ ID NO: 7), and M1 (SEQ ID NO: 9) gene each encodes at least one CTL epitope.

5. The T cell-based immunogenic composition of claim 2, wherein the CTL epitope of the NP protein is MASQGTKRSYEQMET (SEQ ID NO: 11), GIGRFYIQMCTELKL (SEQ ID NO: 12), MVLSAFDERRN (SEQ ID NO: 13), YLEEHPSAGKDPKKTGGPIY (SEQ ID NO: 14), LYDKEEIRRIWRQANNG (SEQ ID NO: 15), ATYQRTRAL (SEQ ID NO: 16), YERMCNILKG (SEQ ID NO: 17), QVRESRNPGNAEIEDLIFLA (SEQ ID NO: 18), QLVWMACHSAAFEDLRVSSF (SEQ ID NO: 19), or QPTFSVQRNLPF (SEQ ID NO: 20);the CTL epitope of the PA protein is KIETNKFAAICTHLEVCFMYSDFHF (SEQ ID NO: 21), RTMAWTVVNSI (SEQ ID NO: 22), VEKPKFLPDLY (SEQ ID NO: 23), YYLEKANKIKSE (SEQ ID NO: 24), THIHIFSFTGEEMA (SEQ ID NO: 25), RGLWDSFRQSERGEETIEE (SEQ ID NO: 26), RSKFLLMDALKLSIE (SEQ ID NO: 27), HEGEGIPLYDAIKC (SEQ ID NO: 28), SQLKWALGENMA (SEQ ID NO: 29), EFNKACELTDSSWI (SEQ ID NO: 30), SRPMFLYVRTNGTSK (SEQ ID NO: 31), or AESRKLLLI (SEQ ID NO: 32);the CTL epitope of the PB1 protein is DVNPTLLFLKVPAQNAISTTFPYTGDPPYSHGTGTGYTMDTVNRTHQYSE (SEQ ID NO:33), MAFLEESHPGIFENS (SEQ ID NO: 34), VQQTRVDKLTQGRQTYDWTLNRNQPAATALANTIE (SEQ ID NO: 35), TKKMVTQRTIGKKK (SEQ ID NO: 36), FVETLARSICEKLEQSGL (SEQ ID NO: 37), RMFLAMITYITRNQP (SEQ ID NO: 38), LSIAPIMFSNKMARLGKGYMFESKSMKLRTQIPAEMLA (SEQ ID NO: 39), SPGMMMGMFNMLSTVLGVS (SEQ ID NO: 40), GINMSKKKSYIN (SEQ ID NO: 41), TGTFEFTSFFYRYGFVANFSMELPSFGVSGINESADMSI (SEQ ID NO: 42), GVTVIKNNMINNDLGPATAQMALQLFIKDYRYTYRCHRGDTQIQTRRSFE (SEQ ID NO: 43), VSDGGPNLY (SEQ ID NO: 44), MEYDAVATTHSW (SEQ ID NO: 45), PKRNRSILNTSQRGILEDEQMYQ (SEQ ID NO: 46), or AEIMKICST (SEQ ID NO: 47);the CTL epitope of the PB2 gene is LMSQSRTREILTKTTVDHMAIIKKYTSGRQEKNP (SEQ ID NO: 48), WMMAMKYPI (SEQ ID NO: 49), PERNEQGQTLWSK (SEQ ID NO: 50), PLAVTWWNRNGP (SEQ ID NO:51), GPVHFRNQVKIRR (SEQ ID NO: 52), YIEVLHLTQGTCW (SEQ ID NO: 53), EQMYTPGGEV (SEQ ID NO: 54), NDDVDQSLIIAARNIVRRA (SEQ ID NO: 55), ASLLEMCHSTQIGG (SEQ ID NO: 56), SFSFGGFTFK (SEQ ID NO: 57), LTGNLQTLK (SEQ ID NO: 58), RVHEGYEEFTMVG (SEQ ID NO: 59), RATAILRKATRR (SEQ ID NO: 60), VAMVFSQEDCM (SEQ ID NO: 61), KAVRGDLNF (SEQ ID NO: 62), VNRANQRLNPMHQLLRHFQKDAKVLF (SEQ ID NO: 63), RVSKMGVDEYS (SEQ ID NO: 64), GNVLLSPEEVSETQG (SEQ ID NO: 65), LTITYSSSMMWEINGPESVL (SEQ ID NO: 66), NTYQWIIRNWE (SEQ ID NO: 67), MLYNKMEFEPFQSLVPKA (SEQ ID NO: 68), LGTFDTVQIIKLLPFAAAPP (SEQ ID NO: 69), QSRMQFSSLTVNVRGSGMRILVRGNSPVFNYN (SEQ ID NO: 70), or GVESAVLRGFLI (SEQ ID NO: 71); andthe CTL epitope of the M1 protein is SLLTEVETYVL (SEQ ID NO: 72), KTRPILSPLTKGIL (SEQ ID NO: 73), GFVFTLTVPSE (SEQ ID NO: 74), LYRKLKREITF (SEQ ID NO: 75), ALASCMGLIY (SEQ ID NO: 76), MGTVTTEVAFGLVCA (SEQ ID NO: 77), NRMVLASTTAKAMEQMAGSS (SEQ ID NO: 78), or QARQMVQAMR (SEQ ID NO: 79).

6. The T cell-based immunogenic composition of claim 2, wherein the genes are included in a recombinant virus.

7. The T cell-based immunogenic composition of claim 6, wherein the recombinant virus is a recombinant vaccinia virus, a recombinant adenovirus, a recombinant adeno associated virus, a recombinant retrovirus, or a recombinant lentivirus.

8. The T cell-based immunogenic composition of claim 7, wherein the recombinant virus is the recombinant vaccinia virus.

9. The T cell-based immunogenic composition of claim 7, wherein it is administered through a skin scarification (s.s) route, an intramuscular (i.m.) route, an intranasal (i.n.) route, an intradermal (i.d.) route, an intravenous (i.v.) route, or an intraperitoneal (i.p.) route.

10. The T cell-based immunogenic composition of claim 9, wherein it is administered through the skin scarification (s.s.) route.