The present invention is based on a new fusion partner for various self- or non-self- proteins to induce strong therapeutic antibody responses. By using a short multimerizing fusion partner based on the pentamerizing part of hagfish or lamprey variable leukocyte receptors (VLRs) the antibody titers increase markedly compared to other fusion partners. The invention also relates to the use of these multimerizing regions for any therapeutic target, self or non-self-protein to induce a strong therapeutic or preventive antibody response against the target antigen.

Background of the invention

Vaccines targeting various infectious diseases are one of the major success stories in human and veterinary medicine. Such vaccines have together with antibiotics probably been more important for human and companion animal health than any other part of human or veterinary medicine. Due to their success the interest is now also increasing for using vaccine technology for the treatment of non-infectious diseases as for example allergies, autoimmunity disorders and cancer [ 1 ]. However, the targets for these diseases are in general not foreign molecules but instead self-antigens, which pose problems with efficacy. It is namely considerably more difficult to mount a strong antibody response to a self-antigen compared to a non-self, bacterial, viral or parasite protein, due to tolerance mechanisms.

We have, during a number of years, been working on the development of vaccines against atopic allergies and cancer and then have become interested in deciphering the mechanisms behind self-tolerance and how to safely use the breaking of self-tolerance to treat non-infectious diseases [2- 5]. One key factor in breaking self-tolerance is the form in which the self-antigen is being presented. It is usually very difficult, or almost impossible, to mount an immune response against non-modified self-antigens [6]. However, fusing the self-antigen with a foreign (non-self) protein (fusion partner) has in numerous studies been shown to result in the induction of an anti-self-antibody response. The effect is here most likely a result of the recruitment of T helper cells directed against the non-self- region of the antigen to give help to the few autoreactive B cells that we have in the circulation [ 1 ]. The non-self T cell epitopes are then probably presented to the T cells by the autoreactive B cells. By this mechanism the autoreactive B cells receive the costimulatory signals needed for expansion. We have also observed that we need a potent adjuvant to elicit an immune response strong enough to give a clinical effect. The fact that we both need a potent adjuvant and a modified self-antigen is a very important safety factor. Neither a modified self-protein by itself nor an adjuvant with a non-modified self-antigen will result in an antibody response of any magnitude. To obtain a fusion partner that induces high antibody titers against the self-antigen and at the same time gives minimal side effects, a broad screening for potent candidate proteins was initiated. During these studies we observed that most non-self-proteins are highly immunogenic and induce very strong antibody responses against the non-self-region of the fusion protein. Antibodies directed against this region then result in the clearance of the antigen by immune complex formation and phagocytosis, which reduces the therapeutic effect of the vaccine. However, during this screening process a new fusion partner was identified that by almost all criteria exceeded our expectations. A short region from a hagfish or lamprey immunoglobulin like molecule forms pentamers spontaneously [7-10]. By this multimerization they form clusters where the non-self-part becomes hidden in the central part of the molecule. This markedly reduces the anti-non-self-titers. Using this region as fusion partner also resulted in consistently much higher and more stable titers against the self-antigen than any other non- self-protein tested. This new non-self-fusion partner is therefore a major breakthrough in the field. It induces several fold higher titers of anti-target antibodies than any other fusion partner tested and appears to be very safe to use. One additional very important characteristic of this protein is the possibility to express it as soluble protein, including as a fusion protein for most antigens, in a prokaryotic host, the bacterium E. coli. It also has the advantage of inducing high antibody titers against multiple targets simultaneously, a very important advantage when vaccinating against complex diseases as allergy and cancer.

The Prior Art

A number of different fusion partners have been tested for the development of therapeutic vaccines targeting self-proteins, including Schistozoma Glutathione S-transferase (GST), E. coli Maltose binding protein (MBP), E. coli thioredoxin (Trx), and several viral proteins forming empty capsids. However, none of them have all the positive characteristics harbored by the VLR construct of the present invention, a low molecular weight component that is stably expressed in bacterial cells and that forms multimers spontaneously and thereby hides itself in the interior of the vaccine antigen. This latter characteristic is a major improvement over the prior art. By the multimerization the non-self- protein becomes hidden in the interior of the molecule whereas the self-protein becomes well exposed to the autoreactive B cells. This feature results in relatively low titers against the non-self-antigen that is hidden from the B cells and in high titers against the self-protein. The non-self-fusion partners previously tested, as described above, have also been shown to result in reduced antibody titers against some of the self-antigens when vaccinating against multiple self-targets simultaneously. The new fusion partner of the present invention has been shown to be superior to all other fusion partners previously tested also in this respect. We see consistently high antibody titers against multiple targets when immunizing against them simultaneously in the same individual using VLR. Object of the Invention

The object of the invention is to provide a convenient and cost effective method to treat various inflammatory conditions and cancer. The invention involves the treatment with a vaccine containing a fusion protein consisting of a self or non-self-part constituting the target antigen and the mul- timerizing part of hagfish or lamprey variable leukocyte receptor (VLR) region constituting the fusion partner. Vaccination with such a fusion protein results, when combined with a potent adjuvant, in the induction of high levels of anti-target antibodies and a potent therapeutic effect. The antibody titers are kept low against the non-self-region due to that the non-self-region of the protein is hidden in the interior of the molecule and thereby not can be seen efficiently by the B cells.

Summary of the Invention

The above object is achieved according to the invention by a vaccine formulation, which is characterized by a vaccine target antigen fused to the entire of or any part of the C terminal pentamerizing (multimerizing) region of the hagfish or lamprey variable leukocyte receptor (VLR-A, B or C), in its original or slightly mutated, functionally equivalent form, as a fusion partner. The protein formulation used as vaccine may optionally contain a pharmaceutically acceptable adjuvant.

Legends

Below there are a number of abbreviations used having the following meanings:

VLRB = the (sea lamprey (Petromyzon marinus)) variable lymphocyte receptor B

EDA = the extra domain A of fibronectin

EDB = the extra domain B of fibronectin

TNC-C = the extra domain C of tenascin C

Gal l = the proangiogenic and immunosuppressive protein galectin-1

TRX = the bacterial redox protein thioredoxin

Brief description of the figures

Figure 1 Panel A shows the domain structure of the sea lamprey {Petromyzon marinus) variable lymphocyte receptor B (VLRB) and recombinant self- to non-self-fusion proteins (EDB- VLRB, EDB-EDB-VLRB, EDA-EDA-VLRB, T CC-TNCC-VLRB, EDA-T CC-VLRB, Gal l - VLRB, Gal l -Gal l -VLRB). VLRB consists of a signal peptide (SP), leucine-rich repeat (LRR) domains and a connecting peptide (CP) which is separated via a Threonine/Proline-rich stalk from a cys- teine-rich C-terminal region (black). Cysteines allowing formation of disulfide bridges are shown in grey. The part of the sequence highlighted in bold was taken for design of fusion proteins to self- antigens. A membrane anchorage motif (DGG) was removed from the original sequence by exchang- ing an Asp to Ser (underlined, italics). One or two domains of self-antigen EDB were fused C- terminally to the non-self VLRB-peptide. A His-tag was added to the N-terminus for purification of the proteins. Panel B shows the amino acid sequences of self-to-non-self fusion proteins as illustrated in Panel A.

Figure 2 shows recombinant protein of all three constructs produced in bacteria (E. coli) after separation on a 4-12% gradient SDS-PAGE gel under reducing or non-reducing conditions. Under the non-reducing conditions we see the multimers that form by covalent Cys-Cys crossl inking in the VLRB region of the protein.

Figure 3 Panel A shows the antibody titers against EDB in mice at week 5 of immunization. The mice had then been receiving two injections, at day 1 and 14, of the vaccine antigen mixed in adjuvant and the mice blood was sampled at week 5. Two different antigens were compared for their efficacy, one consisting of a fusion protein between EDB and the bacterial redox protein thi- oredoxin (TRX) and the second protein a fusion protein between EDB and the C terminal part of Lamprey VLRB. As can be seen from the figure the VLRB fusion protein results in a 2-3-fold higher antibody titer compared to the EDB-TRX protein. Panel B shows the corresponding analysis for the EDA-VLRB protein. Also here a marked increase in titers is observed. Panel C shows the corresponding analysis for the Gal 1 -VLRB protein. Here the difference is more than 5 fold in titers when using the VLRB fusion protein over the TRX protein.

Figure 4 Antibody titers against three different self-proteins in mice injected simultaneously (Combination) or just a single protein with the three VLRB fusions; EDB, EDA and TNC-C fusions. As can be seen from the figure all tree VLRB fusion proteins induced as high antibody titers in these animals as when each antigen was injected alone. This in contrast to the TRX fusion proteins that often showed marked inhibition when injected together as previously shown [6]. Panel A shows the result from the analysis of the EDB protein, Panel B the EDA protein and Panel C the TNC-C protein.

Figure 5 shows the ratios between non-self and self-responses when using TRX or VLRB fusion proteins. As can be seen from this panel a much more favorable response with relatively higher anti-self compared to non-self-responses is seen with the VLRB fusions. Description of the invention

A first embodiment relates to a vaccine composition stimulating the immune system to induce a strong therapeutic or preventive antibody response against any therapeutic target self or non- self-antigen or protein and containing the target antigen, or a part or fragment thereof, being fused to a fusion partner, and optionally a pharmaceutically acceptable adjuvant. This fusion partner according to the invention is the entire or any part of the C terminal pentamerizing (multimerizing) region of the hagfish or lamprey variable leukocyte receptor (VLR-A, B or C), in its original or slightly mutated, functionally equivalent form.

The vaccine composition of the inventions can be used for therapeutic or preventive treatment of cancer tumors, allergic inflammation, autoimmune disorders or other inflammatory conditions.

Preferably, the vaccine composition is directed against self-proteins expressed in and around tumor vessels and containing a single or a combination of the amino acid sequences of the extra domain B of fibronectin (EDB), the extra domain A of fibronectin (EDA) or the extra domain C of tenascin C (TNC-C) or at least one fragment thereof, in its original or multimerized form, coupled to the fusion partner according to the invention.

Also preferably, the vaccine compositions is directed against self-proteins acting as immunosupressants in treating tumors and containing the proangiogenic and immunosuppressive protein galectin-1 or at least one fragment thereof, in its original or multimerized form, coupled to the fusion partner according to the invention.

Further preferably, the vaccine compositions is directed against self-proteins acting as early regulators of inflammatory conditions and containing a single or combinations of the cytokines 1L-33, 11-31 , IL- 1 8, IL-17 and thymic stromal lymphopoietin (TSLP) or at least one fragment thereof, in its original or multimerized form, coupled to the fusion partner according to the invention.

The vaccine compositions of the invention may also contain further additives necces- sary for the formulation of a stable and pharmaceutically acceptable vaccine, such as preservatives.

A second embodiment of the invention relates to the very fusion partner to generate a vaccine antigen to target various self or non-self-proteins in a preventive or therapeutic vaccine composition, the fusion partner being the entire or any part of the C terminal pentamerizing (multimeriz- ing) region of the hagfish or lamprey variable leukocyte receptor (VLR-A, B or C), in its original or slightly mutated, functionally equivalent form.

A third embodiment of the invention relates to the use of the entire or any part of the C terminal pentamerizing (multimerizing) region of the hagfish or lamprey variable leukocyte receptor (VLR-A, B or C), in its original or slightly mutated, functionally equivalent form, as a fusion partner to generate a vaccine antigen to target various self or non-self-proteins in a preventive or therapeutic vaccine.

A fourth embodiment of the invention relates to a method of inhibiting tumor growth, metastasis or inducing stable disease in a patient, or of preventing or treating inflammatory conditions or autoimmune disorders, this being done by administering to a patient in need thereof an effective amount of a vaccine composition according to the invention.

A fifth embodiment of the invention relates to the use of the entire or any part of the C terminal pentamerizing (multimerizing) region of the hagfish or lamprey variable leukocyte receptor (VLR-A, B or C), in its original or slightly mutated, functionally equivalent form, as a fusion partner to generate a vaccine antigen for the treatment of allergic, autoimmune or other inflammatory conditions. Potential targets of interest is here various inflammatory cytokines such as interleukin 33 (IL- 33), 11-3 1 , IL-18, IL-17 and thymic stromal lymphopoietin (TSLP). However, other targets may also be of interest.

By "slightly mutated, functionally equivalent form" is meant any slight routine mutation still preserving the essential structure of the recited region and still upholding the desired effect of enhancing the immunological response, i.e. any slightly mutated form being functionally equivalent to the claimed region".

The vaccine formulation optionally contains a pharmaceutically acceptable adjuvant, like Montanide 1SA720 (which contains natural metabolizable oil and a highly-refined emulsifier), in combination with phosphorothioate-stabilized CpG oligonucleotides. The combination of Montanide ISA 720 and CpG oligonucleotides has been shown to be very potent in mice and rats for a number of different target molecules, including IgE and various cytokines that act as targets for allergy vaccines and different vascular targets that are used for targeting solid tumors. The squalene-based biodegradable adjuvant Montanide ISA 720 has been found to be efficient as a depot adjuvant and very well tolerated in both mice and rats. However, other adjuvants may also be of interest. Example

As indicated above, an antibody response against both the self and the non-self-part of the fusion protein can be induced by injection of a fusion protein between a self and a foreign protein. This is here shown by the vaccination against three tumor vessel specific targets, namely extra domain A and B (EDA and EDB) of fibronectin and extra domain C of tenascin C (TNC-C), and the proangi- ogenic and immunosuppressive protein galectin-1 (Gal l ).

Fusion proteins between the above four self-antigens and a fusion partner, TRX (bacterial redox protein thioredoxin) or the C-terminal part of Lamprey VLRB, were produced in a bacterial host, E. coli, purified and analyzed on an SDS PAGE gel (Figure 2). Only the VLRB fusion proteins form multimers as shown in the non-reducing lanes in Figure 2. The various fusion proteins were then mixed with a depot adjuvant, Montanide ISA720 and a CpG containing oligonucleotide, and injected into the test animals. The animals were given two injections, day 1 and week 3. Blood samples were taken at day 0 and week 5 and tested for the titers of anti-EDB, -EDA, -TNC-C and -Gal 1 antibodies. As can be seen from Figure 3 the titers induced by the VLRB fusion proteins by far exceed the levels induced by the TRX fusion proteins. An approximately 3-5 fold increase in titers is observed.

The ratio of non-self vs self-antibodies is also lower for the VLRB fusions as seen from Figure 5. The shielding of the VLRB in the interior of the molecule and the smaller size of this fusion partner compared to the TRX is here most likely the explanation for this effect. Both the shielding and the smaller size result in a slower rate of antigen clearance by immune complex formation and uptake by phagocytic cells.

The effect after immunizing with several targets simultaneously was also studied using the three tumor vessel targets EDB, EDA and TNC-C fused to TRX or VLRB. As can be seen from Figure 4 the antibody titers were high for all three antigens when immunizing with the VLRB fusion proteins whereas the titers for EDA and TNC-C were very low compared to the EDB titers when immunizing with the three TRX fusion proteins (previously shown, reference 6). The high titers obtained of several targets in animals injected with them simultaneously is a major advantage with the VLRB fusion proteins compared to the TRX fusions as representing prior art. References

1 . Hellman L (2008) Therapeutic vaccines against IgE-mediated allergies. Expert Rev Vaccines 7:

193-208.

2. Hellman L (1994) Profound reduction in allergen sensitivity following treatment with a novel allergy vaccine. Eur J Immunol 24: 415-420.

3. Johansson J, Ledin A, Vernersson M, Lovgren-Bengtsson , Hellman L (2004) Identification of adjuvants that enhance the therapeutic antibody response to host IgE. Vaccine 22: 2873-2880.

4. Johansson J, Hellman L (2007) Modifications increasing the efficacy of recombinant vaccines; marked increase in antibody titers with moderately repetitive variants of a therapeutic allergy vaccine. Vaccine 25: 1676- 1682.

5. Ringvall M, Huijbers EJ, Ahooghalandari Ρ, Alekseeva L, Andronova T, et al. (2009) Identification of potent biodegradable adjuvants that efficiently break self-tolerance~a key issue in the development of therapeutic vaccines. Vaccine 28: 48-52.

6. Saupe F, Huijbers EJ, Hein T, Femel J, Cedervall J, et al. (2015) Vaccines targeting self-antigens: mechanisms and efficacy-determining parameters. FASEB J 29: 3253-3262.

7. Pancer Z, Amemiya CT, Ehrhardt GR, Ceitlin J, Gartland GL, et al. (2004) Somatic diversification of variable lymphocyte receptors in the agnathan sea lamprey. Nature 430: 174-180.

8. Alder MN, Herrin BR, Sadlonova A, Stockard CR, Grizzle WE, et al. (2008) Antibody responses of variable lymphocyte receptors in the lamprey. Nat Immunol 9: 319-327.

9. Altman MO, Bennink JR, Yewdell JW, Herrin BR (2015) Lamprey VLRB response to influenza virus supports universal rules of immunogenicity and antigenicity. Elife 4.

10. Kasahara M, Sutoh Y (2014) Two forms of adaptive immunity in vertebrates: similarities and differences. Adv Immunol 122: 59-90.