Polypeptides containing mutant forms of human VEGF-A with disulfide bond rearrangements and compositions containing the same

Abstract

Polypeptides comprising functional variants of human vascular endothelial growth factor A (VEGF-A) isoforms folded in a non-native rearrangement, in which the second and fourth cysteines of the variant polypeptide chain form only intramolecular bridges, and the seventh and eighth cysteines are only part of intermolecular bonds. The present invention further includes antigen preparations containing at least one of these polypeptides, as well as pharmaceutical compositions comprising such antigen preparations and vaccine adjuvants. Antigen preparations according to the present invention are used in the manufacture of drugs for the treatment of diseases associated with increased angiogenesis, inflammation, and immunosuppression, and for the restoration of the immune system.

Description

[Technical Field]

[0001] The present invention relates to the fields of biotechnology and human health. The present invention provides polypeptides comprising functional variants of human vascular endothelial growth factor A (VEGF-A) folded with a non-native arrangement of disulfide bridges. The present invention provides the basis for producing compositions comprising these polypeptides for use in the prevention and treatment of pathologies that progress through increased angiogenesis, inflammation, and immunosuppression. [Background technology]

[0002] The VEGF-A system and its receptors are molecular complexes whose interactions specifically regulate endothelial cell proliferation, permeability, plasticity, and motility, resulting in positive effects on pathological angiogenesis. Dysregulation of VEGF-A expression and its receptors occurs in both tumor cells and stroma, where VEGF-A is specifically expressed and regulates endothelial cell function. The paracrine effects of tumor VEGF-A on peritumoral endothelial cells, as well as the autocrine effects of this growth factor on tumor cells and stroma expressing its receptors, have been reported primarily over the past 20 years (Carmeliet, Nat Med, 2003:9:653-60; Mashreghi, et al., J Cell Physiol, 2018:233:2949-65).

[0003] VEGF-A induces increased proliferation, motility, and organization in endothelial cells, resulting in the formation of new blood vessels whose configuration and maturation depend on the concentration gradient of VEGF-A and its isoforms, as well as the presence of other proangiogenic factors and their receptors (Carmeliet, Nat Med, 2003:9:653-60). In addition to this primary function, some evidence regarding the expression of type 1 and type 2 VEGF-A receptors and co-receptors in cells of myeloid origin points to this factor as a key mediator in inflammatory and immunosuppressive processes. For example, it is known that VEGF-A interacts with its type 1 receptor in dendritic cells and reduces their maturation via NFκB (Gabrilovich, et al., Nat Med, 1996:2:1096-103; Gabrilovich, et al., Blood, 1998:92:4150-66). Furthermore, VEGF-A binds to its type 2 receptor (VEGFR2) induced on effector T cells as a regulatory mechanism to inhibit interferon-gamma (IFN-γ) secretion (Ziogas, et al., Int J Cancer, 2012:130:857-64). VEGF-A is also an essential mediator of inflammatory processes, inducing phenotypic changes in macrophages and neutrophils in the lesion microenvironment, particularly in association with neoplastic and arthritic phenomena (Voron, et al., Front Oncol, 2014:4:70).

[0004] In this context, the VEGF-A protein family and their receptors are targets for active and passive immunotherapy in diseases that progress through previously described processes. The use of VEGF-A and its receptors as targets has been validated by the approval of passive immunotherapy with Avastin, sorafenib, and sunitinib as first-line treatments for multiple tumor types and age-related macular degeneration (AMD) (Wentink, et al., Biochim Biophys Acta, 2015:1855:155-71). Active immunotherapy with this growth factor, which is involved in inducing autoimmune responses, has developed more slowly. Until 2002, only a few studies using VEGF-VEGFR as targets had been reported. Most of these efforts have been devoted to studying the immunogenicity, antiangiogenic, and antitumor effects of heterologous variants (Wei, et al., Proc Natl Acad Sci USA, 2001:98:11545-50) or proteins with high structural and functional homology to VEGFA (WO 99 / 45018 and WO 00 / 53219). The use of heterologous variants induces high neutralizing and specific antibody titers in the absence of cellular responses, whereas the use of homologous molecules results in a modest immune response without any evidence of antitumor or antimetastatic effects. None of these strategies has been translated into clinical practice.

[0005] The use of functional variants of VEGF-A in active immunization specific for this factor was described in 2002 (patent application no. PCT / CU03 / 00004), using either naked deoxyribonucleic acid (DNA) or recombinant protein fused to immunostimulatory sequences used as adjuvants.

[0006] Administration of VEGF-A mutants (Arg82, Lys84, His86 → Ala82, Ala84, Ala86 or Arg82, Lys84, His86 → Glu82, Glu84, Glu86) resulted in the induction of T-specific antitumor responses when used with naked DNA-based immunization (Bequet-Romero, et al., Angiogenesis, 2007:10:23-34) or protein plus adjuvant-based immunization (Morera, et al., Angiogenesis, 2008:11:381-93). Protein-based formulations induced VEGF-A-specific antibodies that inhibited its binding to VEGFR1 and VEGFR2 (Morera, et al., Angiogenesis, 2008:11:381-93; Morera, et al., Vaccine, 2012:30:368-77). Despite mutations introduced into VEGF-A that disrupt its protein binding to VEGFR-2, it was possible to induce antibodies that neutralize this interaction. The use of this vaccine strategy resulted in relevant antitumor and antimetastatic effects in mouse models of melanoma (MB16F10), lung cancer (3LLD122 y TC1), breast cancer (F3II), and colorectal cancer (CT26). This mutant demonstrated a direct cellular response against syngeneic tumor cells, accompanied by the secretion of IFN-γ after incubation with VEGF-A (Bequet-Romero, et al., Vaccine, 2012:30:1790-9). Unlike other strategies, this one uses a recombinant antigen produced in bacteria that represents VEGF-A isoform 121, in which the cysteine ​​residues involved in VEGF-A dimer formation have not been removed. The use of this strategy in the species studied demonstrated the presence of a potential reserve of superior immune responses (Sanchez Ramirez, et al., BMC Immunol, 2017:18:39).

[0007] By using strategies such as varying adjuvants and their concentrations, as well as altering the amount of antigen administered, higher immune responses have been achieved in preclinical models and in the clinic, with concurrent benefits (Morera, et al., Angiogenesis, 2008:11:381-93; Gavilondo, et al., Vaccine, 2014:32:2241-50; Perez Sanchez, et al., Hum Vaccin Immunother, 2015:11:2030-7; Sanchez Ramirez, et al., BMC Immunol, 2017:18:39). These higher immune responses occur without compromising the physiological parameters of the species studied, while maintaining the autoregulated nature of the response. Antibody levels obtained under this immunization schedule were significantly lower than those achieved after bolus administration of therapeutic antibodies. The factors previously discussed indicate that there remains potential for improving vaccine responses without inducing adverse effects.

[0008] therefore, It is important to obtain variants of human VEGF-A that have better antigenic potential when administered as part of a vaccine preparation, which may be Anti-angiogenic and anti-tumor effects , anti Inflammatory effects, and enhanced immunity repair effect induces a significantly superior immune response that can be translated into . Summary of the Invention [Means for solving the problem]

[0009] The present invention solves the aforementioned problems by providing polypeptides comprising functional variants of human VEGF-A isoforms folded with a non-native rearrangement of disulfide bridges, wherein the second and fourth cysteines form only intramolecular bonds, and the seventh and eighth cysteines of the same molecule are found only as part of intermolecular bridges.

[0010] In the present invention, a functional variant of VEGF-A is a molecule that has 95% sequence identity with the naturally occurring variant, but differs from the former by the fact that it does not bind to VEGFR2 and induce the signaling associated with receptor binding.

[0011] The polypeptides described herein are related to human VEGF proteins, particularly VEGF-A and its isoforms, and have not been previously described. They were primarily produced by modifying the purification process of the polypeptide defined by SEQ ID NO: 2 (Morera, et al., Angiogenesis, 2008:11:381-93). Analysis of these polypeptides showed increased stability, immunogenicity, and antitumor efficacy compared to the original antigen preparation PVM (Examples 1, 2, 3, 4, and 5).

[0012] To obtain a polypeptide related to the present invention comprising a functional variant of a human VEGF-A isoform, amino acids involved in VEGFR2 binding are mutated. In one embodiment of the present invention, the polypeptide is a VEGF-A 121 , VEGF-A 145 , VEGF-A 165 , VEGF-A 189 and VEGF-A 206 The immunogenicity-increasing cysteine ​​configuration claimed in the present invention is characterized in that the human VEGF-A isoform is selected from the group comprising: 121 These isoforms share the canonical cysteine ​​structure described for VEGF-A (Figure 1) and exhibit the same therapeutic effects. Therefore, the same cloning and expression system can be used to identify VEGF-A 145 , VEGF-A 165 , VEGF-A 189 and VEGF-A 206 A sequence corresponding to the following was inserted: Comparative evaluation between products that conserve the cysteine ​​knot and those that do not demonstrate that in the absence of this canonical structure, the increased immunogenicity described for PVM and PVM-I can be replicated (Example 2).

[0013] In other embodiments, a polypeptide comprising a functional variant of a human VEGF isoform comprises a rearrangement of the intramolecular disulfide bridge between the first and second cysteines, the third and fourth cysteines, and the fifth and sixth cysteines, and an intermolecular bond is formed between the seventh cysteine ​​of two polypeptide chains, the eighth cysteine ​​of two polypeptide chains, and the last cysteine ​​of two polypeptide chains.

[0014] In other embodiments, a polypeptide comprising a functional variant of a human VEGF isoform comprises a rearrangement of intramolecular disulfide bridges between the first and second cysteines, the third and fourth cysteines, and the fifth and sixth cysteines, and an intermolecular bond is formed between the seventh and eighth cysteines of two different polypeptide chains and the last cysteines of two polypeptide chains.

[0015] In other embodiments, a polypeptide comprising a functional variant of a human VEGF isoform comprises a rearrangement of the intramolecular disulfide bridges between the first and second cysteines, the third and fifth cysteines, and the fourth and sixth cysteines, and an intermolecular bond is formed between the seventh cysteine ​​of two polypeptide chains, the eighth cysteine ​​of two polypeptide chains, and the last cysteine ​​of two polypeptide chains.

[0016] In other embodiments, a polypeptide comprising a functional variant of a human VEGF isoform comprises a rearrangement of intramolecular disulfide bridges between the first and second cysteines, the third and fifth cysteines, and the fourth and sixth cysteines, and an intermolecular bond is formed between the seventh and eighth cysteines of two different polypeptide chains and the last cysteines of two polypeptide chains.

[0017] In other embodiments, a polypeptide comprising a functional variant of a human VEGF isoform comprises a rearrangement of the intramolecular disulfide bridges between the first and second cysteines, the third and sixth cysteines, and the fourth and fifth cysteines, and an intermolecular bond is formed between the seventh cysteine ​​of two polypeptide chains, the eighth cysteine ​​of two polypeptide chains, and the last cysteine ​​of two polypeptide chains.

[0018] In other embodiments, the polypeptide comprising a functional variant of a human VEGF isoform comprises a rearrangement of the intramolecular disulfide bridges between the first and second cysteines, the third and sixth cysteines, and the fourth and fifth cysteines, and an intermolecular bond is formed between the seventh and eighth cysteines of two different polypeptide chains and the last cysteines of two polypeptide chains.

[0019] In other embodiments, the polypeptide comprising a functional variant of a human VEGF isoform comprises a rearrangement of the intramolecular disulfide bridge between the first and last cysteines, the second and third cysteines, and the fourth and fifth cysteines, and an intermolecular bond is formed between the seventh cysteines of the two polypeptide chains and the eighth cysteines of the two polypeptide chains.

[0020] In other embodiments, the polypeptide comprising a functional variant of a human VEGF isoform comprises a rearrangement of the intramolecular disulfide bridge between the first and last cysteines, the second and third cysteines, and the fourth and fifth cysteines, and an intermolecular bond is formed between the seventh and eighth cysteines of two different polypeptide chains.

[0021] In other embodiments, the polypeptide comprising a functional variant of a human VEGF isoform comprises a rearrangement of the intramolecular disulfide bridge between the first and last cysteines, the second and fourth cysteines, and the third and fifth cysteines, and an intermolecular bond is formed between the seventh cysteines of the two polypeptide chains and the eighth cysteines of the two polypeptide chains.

[0022] In other embodiments, the polypeptide comprising a functional variant of a human VEGF isoform comprises a rearrangement of the intramolecular disulfide bridge between the first and last cysteines, the second and fourth cysteines, and the third and fifth cysteines, and an intermolecular bond is formed between the seventh and eighth cysteines of two different polypeptide chains.

[0023] In one embodiment, a polypeptide comprising a functional variant of a human VEGF isoform comprises a rearrangement of the intramolecular disulfide bridge between the first and last cysteines, the second and fifth cysteines, and the third and fourth cysteines, and an intermolecular bond is formed between the seventh cysteines of the two polypeptide chains and the eighth cysteines of the two polypeptide chains.

[0024] In other embodiments, the polypeptide comprising a functional variant of a human VEGF isoform comprises a rearrangement of the intramolecular disulfide bridge between the first and last cysteines, the second and fifth cysteines, and the third and fourth cysteines, and an intermolecular bond is formed between the seventh and eighth cysteines of two different polypeptide chains.

[0025] In certain embodiments of the present invention, the polypeptide comprising a functional variant of a human VEGF-A isoform further comprises an amino-terminal segment that increases its expression in bacteria and a carboxy-terminal fragment that facilitates the purification process.

[0026] Without limiting the scope of the present invention, a preferred embodiment uses a recombinant protein isomer mixture containing 46 amino acids of the amino-terminal fragment of protein P64K (SEQ ID NO: 24) from Neisseria meningitidis. In this chimeric protein, the amino-terminal segment and the amino-terminal portion of VEGF isoform 121 are separated by 13 amino acids connecting these two segments. Furthermore, the protein has a carboxy-terminal six-histidine sequence useful for antigen production. The conservation of Cys110 and 119, which correspond to Cys51 and 60 in native VEGF-A (Figure 1), in the fusion polypeptide may promote the formation of several structural isomers accompanied by changes in the secondary structure of the protein. Previous studies by Bequet-Romero et al. and Morera et al. only used VEGF-A mutants containing these cysteines. Nevertheless, in both of them, an intramolecular disulfide bridge was detected between cysteine ​​2 and cysteine ​​4 in the polypeptide chain. The presence of these two Cys residues and the process for antigen production in which Cys2 and Cys4 form only intramolecular bonds constitute a new approach to protein production for vaccine purposes. Overall, this strategy can expose sites that enhance immunological responses. The former may be due to recognition by antibodies that neutralize the binding of VEGF-A to its type 2 receptor, or to the fact that antigen digestion by the proteasome is promoted, thus allowing new peptides to be presented by antigen-presenting cells.

[0027] A specific embodiment of the present invention is a polypeptide having an amino acid sequence comprising SEQ ID NO: 18 to SEQ ID NO: 23. These polypeptides have increased immunogenic properties compared to polypeptides that preserve the tertiary structure described for VEGF-A isoform 121. Modifications in the tertiary structure were detected in these polypeptides, which explain the desired superior effect. This superior immunogenicity was surprising, since changes in secondary structure have previously been described as not optimal for generating an effective immune response to human VEGF-A (Wentink, et al., Proc Natl Acad Sci USA, 2016:113:12532-7).

[0028] The purification process established for the polypeptide identified by SEQ ID NO:2 produces structural isomers of VEGF-A with a Cys-knot structure similar to that of the wild-type molecule, as shown in Example 1. Unexpectedly, modifications to the purification process result in enrichment of a family of isomers displaying an intramolecular disulfide bond involving Cys110 and 119, corresponding to Cys51 and 60 from VEGF-A that naturally form intermolecular bridges. In the antigen preparation designated herein as PVM-I, other intermolecular bonds involving Cys161, 163, and 175 were observed, a fact related to the production of oligomeric structures stabilized by disulfide bridges. There is no prior art information regarding the formation, stabilization, and sequence of the folds described herein. This is related to the fact that virtually all studies dealing with recombinant production of VEGF-A have preserved the sequence of Cys residues corresponding to Cys51 and 60 in native VEGF-A. When these cysteines were conserved in the primary sequence, a renaturalization process was used to regenerate the native disulfide bridges (Pizarro, et al., Protein Expression and Purification, 2010:72:184-93). Thus, the detection of novel non-canonical linkages in the absence of native linkages and their relationship to increased stability and biological activity of PVM-I preparations is an unexpected and surprising finding of the present invention.

[0029] The present invention is not limited to a specific format for producing the structural isomers present in the antigen preparation PVM-I. As shown in Example 7, these isomers can be obtained and separated using other procedures. Similarly, this example demonstrates the equal contribution of the identified isomers to increased immunogenicity. Several structural isoforms coexist within protein fractions that are less resistant to trypsin-based digestion, and they exhibit similar potential for enhancing specific immune responses.

[0030] PVM-I preparations consist of either isomeric mixtures or specific isomers isolated using reverse-phase chromatography, which can efficiently separate the 12 isomeric variants. Administration of the isolated isomers reproduces the immunological effects observed with preparations containing all of them, making any of the isomeric mixtures potentially effective (Example 7). This may be related to the fact that in all cases, peptide generation by endoproteases such as trypsin is favored, and trypsin is one of the essential components of the proteasome system responsible for generating peptides presented to the immune system within the framework of MHC molecules (Pamer, et al., Annu Rev Immunology 1998:323-358).

[0031] The present invention also relates to an antigen preparation comprising at least one polypeptide of a functional variant of a VEGF-A isoform folded with a non-native rearrangement of disulfide bridges, in which the second and fourth cysteines of the polypeptide chain form only intramolecular bonds and the seventh and eighth cysteines of the polypeptide chain form only intermolecular bridges. The antigen preparation also comprises at least one pharmaceutically acceptable excipient or diluent. The immunogenic dose can be administered in a vehicle approved for pharmaceutical use that is nontoxic and has no inherent therapeutic effect. These vehicles include ion exchangers, alum, aluminum stearate, lecithin, serum proteins, buffer substances, glycine, sorbic acid, potassium sorbate, partial glyceride mixtures of fatty acids of plant origin, water, salts and electrolytes, polyvinylpyrrolidone, cellulose, and polyethylene glycol-based substances. In the present invention, Tris chloride buffer is preferentially used as a vehicle for vaccine preparations.

[0032] Since the primary object of the present invention is polypeptides with secondary / tertiary structures that differ significantly from those found in nature, the present invention is not limited to antigen preparations obtained by a given procedure.

[0033] Isomeric families composed of polypeptides containing variants of VEGF-A isoforms can be obtained in homogeneous compositions using other strategies, including, inter alia, purification of structural isomers using reverse-phase high-pressure liquid chromatography (RP-HPLC) or controlled regeneration with imbalanced denaturants. As shown in Example 7, structural isoforms can be separated using RP-HPLC, and their independent use at the same concentration produces similar immunogenicity, antitumor, and antimetastatic effects compared to the previously described mixture (PVM-I).

[0034] The PVM-I preparation is an antigen variant with higher immunogenicity and stability based on preliminary studies using Freund's adjuvant (Example 1). These studies were extended to adjuvants and schemes relevant to therapeutic scenarios in non-human mammals (Examples 2-5, and 9-13) and humans (Examples 14 and 15). While the use of more stable antigen variants is relevant in production and commercial situations, the increased immunogenicity provides a superior therapeutic solution (Examples 2-5, 9-15).

[0035] In a preferred embodiment, the present invention also describes a pharmaceutical composition comprising a functional variant of human VEGF-A that folds with a non-native rearrangement of disulfide bridges, wherein the second and fourth cysteines of the polypeptide sequence form only intramolecular bonds and the seventh and eighth cysteines form only intermolecular bridges.

[0036] The compositions of the present invention contain at least a pharmaceutically acceptable adjuvant. To enhance immune responses, the structural isomers described in the present invention can be combined with previously described immunostimulants. These include inorganic salts, immunostimulants such as cytokines, molecular adjuvants (CD40, CD154, MHC type I invariant chain, LFA3), saponins, muramyl dipeptide derivatives, CpG oligonucleotides, lipopolysaccharides, monophosphoryl lipid A and polyphosphatases, lipid particles (i.e., Freund's adjuvant, MF59, and Montanide), liposomes, nanoparticles, virosomes, ISCOMS, coiled substances, microparticle adjuvants, poloxamers, viral and bacterial antigens, and mucosal adjuvants. In certain embodiments, the adjuvant is selected from the group consisting of oil-based adjuvants, aluminum salts, proteoliposomes, and proteoliposomes conjugated to gangliosides.

[0037] Administration of antigens with pharmaceutically acceptable adjuvants, with or without other antitumor agents, as first- or second-line treatment, can contribute to preventing the appearance of new metastases and reducing or even eliminating primary tumors (Examples 2-5, 9-12, and 14). This type of immunotherapy can also be useful in the treatment of, inter alia, acute and chronic inflammatory processes (asthma, shortness of breath, endometriosis, atherosclerosis, tissue edema), infectious diseases (hepatitis, Kaposi's sarcoma), autoimmune diseases (diabetic psoriasis, rheumatoid arthritis), diabetic retinopathy, macular degeneration, neovascular glaucoma, hemangiomas, and angiofibromas (Examples 12, 13, and 15).

[0038] In particular, the present invention describes the administration of vaccine antigens on a weekly schedule when formulated into a group of adjuvants designated NAcGM3-VSSP. These adjuvants can be obtained by conjugating natural or synthetic forms of ganglioside N-acetyl GM3 to outer membrane vesicles derived from N. meningitidis (U.S. Patent No. 6,149,921; International Publication No. 201986056; Regalado, et al., Organic Process Research & Development 2013:17:53-60). Herein, these adjuvant variants, also known as VSSPs (ultrasmall proteoliposomes), are referred to as NAcGM3-VSSP. Those incorporating synthetic N-acetyl GM3 differ in the length of the fatty acid attached to the ganglioside. In general, the use of stearic acid-containing (sNAcGM3-VSSP) or oleic acid-containing (oNAcGM3-VSSP) variants reproduced the results obtained with natural ganglioside variants in terms of immunogenicity, antitumor activity, and antimetastatic activity. In all of these formulations, when PVM-I was used as the antigen, superior antitumor, antimetastatic, anti-inflammatory, and immune-restorative effects were obtained compared to PVM (Examples 4–12).

[0039] Histopathological studies revealed in all cases a significant trend toward a decreased number of functional blood vessels in the primary tumor, decreased tumor cell density, and a decreased mitosis / apoptosis ratio, which correlated inversely with tumor growth. Given that cancer-related lethality is primarily associated with metastatic events, the antigen preparations were further evaluated in aggressive models of spontaneous and experimental metastasis to the lung. In all cases, the PVM-I-based preparations demonstrated significantly better anti-metastatic profiles than those obtained with the vaccine antigen PVM.

[0040] Analysis of metastatic foci in treated animals showed a decrease in vascularity and mitotic / apoptotic balance, parallel to an increase in the number of necrotic foci correlating with the number of metastases found, characteristics similar to those described for treated primary tumors. Interestingly, vaccine-based treatment was not only able to reduce the number of metastases, but also their size and proliferation potential, demonstrating a dual effect of therapeutic intervention on the metastatic process: the implantation of foci and their growth (Example 9).

[0041] Notably, the antigen preparation of the present invention, when formulated in aluminum phosphate, was administered every two weeks, and this adjuvant provided superior antitumor, antimetastatic, anti-inflammatory, and immune-restorative effects compared to PVM-I when used as the antigen.

[0042] At similar concentrations and across all tested adjuvants, the PVM-I preparation was significantly superior to PVM in terms of the specific antibody responses achieved and serum neutralizing capacity. These effects were observed in both mouse strains used, demonstrating the broad spectrum of responses achievable by treatment in diverse haplotype contexts. Similarly, for NAcGM3-VSSP-based formulations, PVM-I is a superior antigen choice for inducing cellular responses that directly eliminate syngeneic tumor cells. The novel antigen preparation also demonstrated greater immunogenicity when assayed in species that share greater homology with native VEGF-A, such as human primates.

[0043] For therapeutic applications, the vaccines from the present invention are administered to mammals, preferably humans, in pharmaceutically acceptable doses using routes known to those skilled in the art. The described antigen preparations can be administered on a simultaneous or sequential schedule with other treatments.

[0044] Administration of the novel antigen preparation PVM-I significantly reduced the growth of tumors of diverse origins, including melanoma, lung cancer, breast cancer, and colon cancer. These results, obtained in tumor models relevant to the translation of other antitumor therapies into clinical practice, demonstrate the applicability of this vaccine strategy to the clinical setting in cancer treatment. Increasing the antigen dose is associated with increased biological effects related to the immune response; therefore, the amount of antigen used for each application depends on the desired effect.

[0045] The present invention also demonstrates for the first time the use of polypeptides comprising functional variants of human VEGF-A isoforms folded with non-native rearrangements of disulfide bridges in the manufacture of medicaments for treating diseases associated with increased angiogenesis, inflammation, and immunosuppression. In certain embodiments, the disease to be treated is selected from the group comprising cancer, macular degeneration, diabetes, rheumatoid arthritis, and edema.

[0046] The novel antigen preparation PVM-I is useful for active immunotherapy of cancer in humans. Its use in the treatment of neoplastic diseases reproduces the immunological effects described in the preclinical phase. VEGF-A-specific antibodies and T lymphocyte clones secreting IFN-γ in response to VEGF-A stimulation were detected in immunized patients. The presence of this immune response was associated with a significant increase in survival. A long-lasting immune response was observed in the group of patients with a complete response, a rare finding for a cancer vaccine. Furthermore, it is interesting to note that therapeutic success is not dependent on complete removal of VEGF-A from the plasma. This factor may be related to the lack of side effects similar to those described for other therapeutic alternatives targeting the VEGF / VEGFR2 axis in both preclinical and clinical settings. Similarly, by achieving a reduction in circulating VEGF-A and systemic neutralization of VEGF-A's ability to bind to VEGFR2, the novel antigen preparation, administered in a pharmaceutically acceptable adjuvant, can be used to treat a range of conditions associated with excessive angiogenesis.

[0047] The present invention does not limit the use of the antigen preparations disclosed herein to specific diseases, but rather illustrates how effective the treatments are in those settings. Thus, administering the antigen preparations of the present invention in the presence of aluminum phosphate or sNAcGM3-VSSP rescues the immune system of tumor-bearing animals (Example 11). Antigen preparation PVM-I also demonstrated increased anti-inflammatory effects compared to the antigen preparation referred to herein as PVM in the context of a collagen-induced arthritis model in DBA / I mice (Example 12). Similarly, antigen preparation PVM-I had greater anti-angiogenic effects in animal models of corneal injury and in human age-related macular degeneration (Examples 13 and 15). This demonstrates the potential utility of this strategy in the treatment of non-tumor diseases. The reduction in corneal angiogenesis as a result of administration of VEGF-A in this compartment indicates an increase in local levels of VEGF-A, demonstrating the potential for strategies to treat age-related macular degeneration and diabetic retinopathy, diseases that can be controlled by inducing an immune response through active immunization with preparations containing the polypeptides of the present invention.

[0048] In another aspect, the present invention discloses the use of a pharmaceutical composition comprising an antigen preparation containing at least one polypeptide comprising a functional variant of a human VEGF-A isoform that folds into a non-native rearrangement of disulfide bridges in restoring the immune system. The novel antigen preparation not only exhibits better immunogenic properties but also leads to superior immune system restoration. Analysis of suppressor myeloid and regulatory T cells was performed in primary and metastatic lesions of animals treated with the antigen preparation of the present invention. This analysis showed that the number of these suppressor cells decreased when the vaccine preparation was administered with different adjuvants and schedules. The functionality of the suppressor cells, as well as their number at the systemic level, was also affected (Example 10).

[0049] The adjuvant effect of the antigen preparation disclosed in this invention was demonstrated in an experimental model expressing ovalbumin antigen (OVA). Sequential or concomitant administration of the PVM-I vaccine antigen in adjuvant with OVA significantly increased OVA-specific cellular and humoral immune responses. Furthermore, the effect of vaccination with PVM-I on cross-presentation of VEGF-A and OVA in the context of class I presentation molecules was observed (Example 11). [Brief explanation of the drawings]

[0050] [Figure 1A] Sequence alignment of the amino acid sequences derived from native VEGF-A with those present in the antigen preparation PVM in isoforms 121, 145, 165, 206, and 189. Cysteines represented by dark Cs (C) are involved in intramolecular bridges, while those underlined (C) form intermolecular bonds. [Figure 1B] FIG. 1 is a schematic representation of disulfide bond configurations in naturally occurring VEGF-A variants. [Figure 2] FIG. 1 is a schematic representation of the disulfide bonds detected in the antigen preparation PVM-I. [Figure 3] FIG. 1 shows assessment of the presence of the canonical "cystine knot" from VEGF-A using native polyacrylamide gel electrophoresis (in the absence of dithiothreitol or β-mercaptoethanol). [Figure 3A] FIG. 1 shows analysis of PVM, CHO-VEGF-A and VM proteins after trypsin digestion. [Figure 3B] Protein analysis of two lots of PVM-I preparations (lanes 2-5) and PVM as a control (lanes 6-7) after trypsin digestion. Molecular weight markers were run in the lanes marked PM1 and PM2. [Figure 4] FIG. 1 shows the evaluation of humoral and cellular responses in advanced cancer patients immunized with PVM-I administered in NAcGM3-VSSP adjuvant (I) or aluminum phosphate (II). [Figure 4A]FIG. 1 shows VEGF-A specific IgG titers for immunized patients. [Figure 4B] FIG. 1 shows a study of the ability of immunized patient sera to block VEGF-VEGFR2 interaction. [Figure 4C] Figure 1 shows the evaluation of VEGF-specific cellular responses measured by INF-γ ELISPOT. In all cases, values ​​represent the difference found compared to values ​​at the start of treatment. [Figure 5] 1 shows an analysis of the immune response and its impact on survival of patients immunized with the antigen preparation PVM-I. Survival time is shown as a function of the positive immune response detected for groups receiving antigen either in NAcGM3-VSSP adjuvant (I) or aluminum phosphate (II).

[0051] example Example 1. Purification and characterization of PVM and PVM-I antigen preparations. DNA encoding human VEGF-A isoform 121, in which amino acids 82, 84, and 86 were mutated by glutamic acid, was cloned into vector PM238 (Morera, et al., Angiogenesis, 2008:11:381-93). The DNA sequence was 100% confirmed and designated SEQ ID NO:1. In this gene construct, the ampicillin resistance gene was interrupted by a gene corresponding to kanamycin resistance without altering expression levels. The plasmid was transformed into Escherichia coli strain BL21, and transformants with the highest expression levels were selected in a chemically defined medium designed to maximize recombinant protein expression in the absence of components of animal origin. The protein was purified according to the procedure described by Morera et al. (Morera, et al., Angiogenesis, 2008:11:381-93). Briefly, proteins were extracted in 50 mM NaHPO buffer, 300 mM NaCl, 6 M urea, pH 7.8 for 16 h at 4 °C and purified according to the manufacturer's instructions (QIAGEN). The buffer was changed to 10 mM Tris, pH 7.4 for size exclusion chromatography on matrix G25 (GE Healthcare).

[0052] The resulting protein preparation was evaluated by mass spectrometry to verify its amino acid sequence. ESI-MS and ESI-MS / MS (electrospray ionization mass spectrometry and electrospray ionization tandem mass spectrometry) were obtained on a QTOF-2 orthogonal hybrid configuration spectrometer (Micromass, UK) equipped with a Z-spray electrospray ionization source (nanoESI). The molecular weight of the reduced protein preparation (21,569.13 Da) corresponded to the theoretical estimate (21,569.31 Da) of human VEGF-A isoform 121, which is fused at the amino terminus to a fragment of the bacterial protein P64K, separated from it by 13 amino acids that serve as a bridge between these polypeptides, and also incorporates a histidine tail at the carboxyl terminus. Mutations introduced into the human VEGF-A sequence were confirmed by detecting glutamic acid instead of arginine, lysine, and histidine at positions 141, 143, and 145, respectively, corresponding to the amino acids in the native VEGF-A sequence.

[0053] The integrity of the amino termini in the samples was verified from the peptide mixture obtained by digestion with Glu-C endoproteinase. 1 VDKRMALVE 9 A signal corresponding to the peptide (doubly charged, theoretical m / z 530.78) was identified. This peptide was sequenced by ESI-MS / MS, and the sequence was identical to that expected for the amino terminus of the protein.

[0054] In the ESI-MS spectrum of the reduced intact protein, fragmentation was induced at the source by increasing the voltage at the input cone of the mass spectrometer, and a signal corresponding to the C-terminus was detected. Analysis of the results was performed to identify peptides after sequencing by ESI-MS / MS. 177 KPRRGSRAHHHHHH 190(doubly charged, m / z 875.46), confirming the carboxyl-terminal sequence. Overall, the sequencing results allowed us to verify the primary structure of the polypeptide present in the antigen preparation (SEQ ID NO: 2). Hereafter, the protein preparation defined by SEQ ID NO: 2 is referred to as PVM.

[0055] Fermentation and purification studies demonstrated superior recovery rates compared to previously described methods (Morera, et al., Angiogenesis, 2008:11:381-93) by modifying growth conditions during fermentation, shortening extraction time, adjusting the specific protein load in nickel affinity chromatography, and incorporating detergents into the wash step. Briefly, during the fermentation process, the fermentation temperature was changed to 28°C during the growth phase, and expression was induced by simply raising the temperature to 37°C without the need for chemical inducers. Protein extraction time from biomass in 6 M urea was also shortened from 16 hours to 2 hours, and a wash step was introduced into nickel affinity chromatography using 0.1% Triton X114. Furthermore, the preparation obtained from molecular exclusion chromatography was formulated in mannitol (40 mg / mL), sucrose (10 mg / mL), and 10 mM Tris-HCl pH 7.4. This final protein preparation is called PVM-I and will be compared below with the preparation obtained in the first process, the PVM preparation.

[0056] In both preparations, the concentration of total protein was assessed using the micro-Coomassie method at dual wavelengths (620 and 450 nm) and percent purity by SDS-polyacrylamide gel electrophoresis (SDS-PAGE). Immunoidentification was performed by immunoblotting using monoclonal antibodies that recognize different segments present in the antigen. In these studies, PVM and PVM-I preparations were identical.

[0057] The conformation and stability of these lyophilized protein preparations were analyzed using analytical-scale molecular exclusion chromatography on a Superdex 200XK10 / 300 column (GE Healthcare). A buffer solution of 10 mM Tris-HCl, 150 mM NaCl, pH 7.4 was used as the mobile phase at a flow rate of 0.5 mL / min. The lyophilized protein preparations were suspended in 1 mL of water, and comparative data showed that upon reconstitution, PVM and PVM-I exhibited identical retention profiles in this type of chromatography. In both cases, soluble aggregates with molecular weights exceeding 670 kDa were formed, according to the molecular weight standards used. However, the reconstituted solution from PVM-I exhibited differential characteristics regarding the stability of the spiked components over time, indicating variation in its structural composition. After reconstitution, the PVM-I preparation maintained its retention profile for 30 days at 4°C in molecular exclusion chromatography, whereas the PVM preparation began to lose significant stability at 72 hours (Table 1). Analysis of the area under the curve of the chromatographic charts shows that the conformational change of over 14% of the PVM protein mass at 72 hours increases significantly at day 30, when the original conformation is lost by over 45%. [Table 1]

[0058] To evaluate the immunogenicity of the PVM and PVM-I protein preparations, the humoral immune response generated in mice was assessed. Two mouse strains (C57Bl / 6 and BALB / c) were used, with 10 animals per group. 100 μg of the protein preparation was administered weekly in a total volume of 250 μL (protein / adjuvant ratio 1:1 v / v) in complete Freund's adjuvant (SIGMA) for the first dose and in incomplete Freund's adjuvant (SIGMA) for the second dose. Serum from the animals was collected 1 week after the second immunization. The specific antibody titers against human VEGF-A present in the serum and their ability to neutralize the interaction of VEGF-A with its type 2 receptor were assessed by ELISA as previously described (Bequet-Romero, et al., Vaccine, 2012:30:1790-9). Titer results are expressed as the maximum dilution at which the presence of specific antibodies to VEGF-A is detected. Neutralizing capacity is expressed as the percentage of maximum VEGF-A / VEGFR2 binding that can be removed by antibodies present in the serum.

[0059] A comparison of immunogenicity demonstrated a significant increase in the level of seroconversion achieved when immunized with the new antigen preparation (PVM-I) compared to the previous one (PVM), even when these equivalent doses were administered, as shown in Table 2. [Table 2]

[0060] Disulfide bridge formation was studied in both preparations, given that (a) the monomers in these protein preparations have identical primary sequences (SEQ ID NO: 2), (b) they contain nine cysteine ​​residues, and (c) the immunogenicity, thermal, and conformational stability of human VEGF-A (which constitutes 63% of the PVM and PVM-I polypeptide sequences) are related to the formation of canonical cysteine ​​structures that have been extensively characterized for the wild-type protein and its family.

[0061] The ordering of disulfide bridges within the protein preparations was previously assessed through trypsin proteolysis studies and analyzed by electrophoresis under non-denaturing conditions and by molecular exclusion chromatography on a Superose 12XK10 / 300 column at 0.5 mL / min using 200 mM Tris-HCl, pH 8.0, as the mobile phase.

[0062] The characteristic arrangement of cysteines in native VEGF-A is known to make it resistant to trypsin digestion (Keck, et al., Arch Biochem Biophys, 1997:344:103-13). Therefore, we compared the digestion of PVM and PVM-I with this enzyme. To this end, the final purification buffer was exchanged into 200 mM Tris-HCl pH 8.0, and the proteins were incubated with or without trypsin at a 50:1 ratio (protein:trypsin) at 37°C for 16 hours. Samples were then taken and the efficiency of trypsin digestion was assessed by non-denaturing native electrophoresis (Figure 3). The following samples were used as controls for the correct conformation of the crosslinks in the VEGF-A molecule: 1) VM protein: human VEGF-A isoform 121 with R80, K82, and H84 mutations relative to E, produced from the periplasm of Escherichia coli (E. coli) (Gavilondo, et al., Vaccine, 2014:32:2241-50), and 2) CHO-VEGF protein: human VEGF-A isoform 121 obtained by transfection of the eukaryotic cell line CHO (Chinese hamster ovary) (Sanchez Ramirez, et al., J Immunoassay Immunochem, 2016:37:636-58). These VEGF-A mutants were purified by a non-denaturing process in the absence of reducing agents. Figure 3 shows how resistant the preparation from the bacterial periplasm was to digestion with trypsin, since 90% of it was recovered in a fraction migrating at a molecular weight lower than that of the undigested protein, consistent with the predicted molecular weight and the presence of a canonical cysteine ​​structure, also known as a "cysteine ​​knot," in the dimeric form of the mutant with or without the P64K segment (19.902 kDa) (Figure 3A).

[0063] Analysis of the PVM preparation showed a similar phenomenon, but this was not observed in the PVM-I preparation, in which peptides corresponding to the cysteine ​​knot did not appear (Figure 3B). Taking this into account, peptides resulting from digestion of the PVM-I preparation with trypsin or GluC were analyzed by mass spectrometry without subjecting the samples to a reduction process, as described for PVM. Table 3 shows the detected disulfide bonds. [Table 3]

[0064] The consistency of the presence of these disulfide bonds was analyzed and verified in a total of eight lots of the final product. The identified structures are found in both the active pharmaceutical ingredient and the final formulation. The specific array of disulfide bridges found for PVM-I differs significantly from that described for the wild-type molecule of VEGF-A and other isoforms of this protein. They also differ from the array described for the protein when obtained by recombinant technology in various prokaryotic and eukaryotic expression systems (Keck, et al., Arch Biochem Biophys, 1997:344:103-13). An extensive search was performed for intermolecular bridges between cysteines 110 and 119, which correspond to the second and fourth cysteines of native VEGF-A. Even though these cysteine ​​residues form intramolecular bonds in the wild-type molecule, peptides corresponding to them were not detected in trypsin- or Glu-C-digested PVM-I preparations under any experimental conditions.

[0065] According to the prior art, the most stable conformation of VEGF-A corresponds to that maintaining the so-called "cysteine ​​knot." However, studies performed with the protein preparations PVM and PVM-I have shown that, under the conditions in which they are purified, higher stability is associated with variants lacking the native conformation. Similarly, despite the data of Timmerman et al. (U.S. Patent Application Publication No. 2012 / 0231000), a comparative analysis of the immunogenicity of PVM-I in Freund's adjuvant compared to PVM indicates that Freund's adjuvant containing the isomer mixture shown in Figure 2 is more immunogenic.

[0066] This study demonstrates that changes in disulfide bond configuration occur in PVM-I preparations compared to PVM, resulting in greater stability and reproducibility following the new protocol for antigen production. Figure 2 shows 12 variants of crosslinks established in stable forms among the nine cysteines detected in the polypeptide sequence of PVM-I.

[0067] Example 2. Comparison of immunogenicity of antigen preparations PVM and PVM-I administered in NAcGM3-VSSP. C56BL / 6 and BALB / c mouse strains, which differ in their ability to generate humoral or cellular immune responses to antigen challenge, were used. In both cases, 100 μg of antigen preparation was administered per dose: a) 100 μg NAcGM3-VSSP incorporating native gangliosides, b) 100 μg sNAcGM3-VSSP incorporating gangliosides with stearic acid, or c) 100 μg oNAcGM3-VSSP incorporating gangliosides with oleic acid. Ten animals per group were immunized weekly for 8 weeks. Animal sera were collected 1 week after each immunization, after the third immunization, and again after the eighth immunization. Antibody titers specific for human VEGF-A and their ability to neutralize the interaction of VEGF-A with its type 2 receptor were assessed by ELISA as described (Bequet-Romero, et al., Vaccine, 2012:30:1790-9).

[0068] Specific IgG titers increased with increasing immunization frequency, reaching a maximum one week after the eighth immunization. The results corresponding to this experimental point are shown in Table 4. As can be seen, for both strains, significantly better results are obtained with the PVM-I antigen preparation after the eighth immunization, both in terms of specific titers and inhibition of VEGF-A binding to its receptor. Comparative analysis of the three variants of the adjuvant NAcGM3-VSSP indicates that the vaccine preparations are equally immunogenic. [Table 4]

[0069] Example 3. Comparison of the immunogenicity of antigen preparations PVM and PVM-I administered in aluminum phosphate. C56BL / 6 and BALB / c mouse strains were used. One hundred micrograms of antigen preparation per dose was administered in 0.7 mg of Al3+ (aluminum phosphate) equivalent. Ten animals per group were immunized every two weeks for a total of four immunizations. Animal serum was collected one week after the second and fourth immunizations. Human VEGF-A-specific antibody titers and their ability to neutralize the interaction between VEGF-A and VEGFR2 were assessed by ELISA as described (Bequet-Romero, et al., Vaccine, 2012:30:1790-9).

[0070] For both antigen types and both mouse strains, increasing titers were observed as the number of administered doses increased. Results for serum samples obtained after the fourth immunization are shown in Table 5. For both strains, after the fourth immunization, significantly better results are obtained for the PVM-I antigen preparation in terms of specific antibody titers and inhibition of VEGF-A binding to its receptor. [Table 5]

[0071] Example 4. Evaluation of cellular responses induced by immunization with antigen preparations of PVM or PVM-I administered in NAcGM3-VSSP. Cellular responses were analyzed as described by Bequet-Romero et al. (Bequet-Romero, et al., Angiogenesis, 2007:10:23-34; Morera, et al., Vaccine, 2012:30:368-77). In this case, responses were evaluated only in the context of the adjuvant NAcGM3-VSSP. Animals (n = 10 per group) were immunized with this adjuvant as described in Example 2 and sacrificed one week after the last immunization. Cells isolated from the spleens were co-incubated with syngeneic tumor cells labeled with CFSE (carboxyfluorescein succinimidyl ester) fluorophore at a 100:1 ratio (effector cells:labeled tumor cells).

[0072] The melanoma cell lines B16F10, EL4 lymphoma, and lung cancer 3LL-D122 syngeneic to the C56BL / 6 strain, as well as the colon cancer lines CT26, breast cancer F3II, and renal RENCA syngeneic to the BALB / c strain, were selected for this study. Flow cytometric analysis of live cells after co-incubation (Space ML-PARTEC) showed that administration of the PVM-I preparation resulted in a stronger and more direct cellular response in both mouse strains, even when similar doses of the two antigen preparations were administered. Table 6 shows the results corresponding to the percentage of tumor cells killed upon encounter with lymphocytes from immunized animals, with the mean value of cells corresponding to vehicle treatment taken as 100% survival. In both mouse strains (C57BL / 6 and BALB / c), administration of the antigen preparation PVM-I (corresponding to group IV in all cases) was superior to the use of the PVM formulation in terms of inducing effective cellular responses. [Table 6]

[0073] Example 5. Efficacy of PVM and PVM-I based immunization for the treatment of solid tumors in mice. In all cases, groups of 12 mice were immunized. The results of immunization with sNAcGM3-VSSP or aluminum phosphate are described for two tumor models. When using the adjuvant sNAcGM3-VSSP, animals were immunized subcutaneously for 8 weeks with a total of 200 μL containing 100 μg of the corresponding antigen and 100 μg of adjuvant. For aluminum phosphate, animals were immunized four times on a biweekly schedule. Each time, 100 μg of antigen was administered in a total volume of 200 μL containing the equivalent of 0.7 mg of Al3+ in the form of aluminum phosphate.

[0074] The melanoma model B16F10 was evaluated in the C57BL / 6 mouse strain. Three days after the fourth immunization (in the presence of sNAcGM3-VSSP) or the second immunization (in the presence of aluminum phosphate), mice were subcutaneously inoculated with a total of 20,000 cells in 100 μL of DMEM culture medium. Table 7 shows the weights of primary tumors surgically removed from euthanized animals 25 days after tumor challenge. [Table 7]

[0075] As can be seen in Table 7, in both adjuvants, treatment of animals with the antigen preparation significantly reduced tumor growth toward day 25 after tumor challenge, but the use of the novel antigen preparation (PVM-I) showed a more significant inhibition. This antitumor response correlated with the presence of both humoral and cellular specific immune responses against VEGF-A (p<0.05, Pearson test).

[0076] Similarly, the antitumor effect was evaluated in the BALB / c mouse strain using the same immunization scheme. Tumor challenge was also performed 3 days after the fourth immunization (in the presence of NAcGM3-VSSP) or the second immunization (in the presence of aluminum phosphate) with 25,000 cells of the CT26 syngeneic colon carcinoma carrying this strain. Comparative analysis of tumor growth 30 days after challenge showed a superior antitumor effect of immunization with the antigen preparation PVM-I compared with the use of the PVM preparation (Table 8). [Table 8]

[0077] As in previous studies, this result correlated with the presence of both humoral and cellular immune responses specific to VEGF-A, which were superior in the case of the use of the antigen preparation PVM-I (p<0.05, Pearson test). In both studies, histological analysis of serial sections of extracted tumors showed antiangiogenic, proapoptotic, and antiproliferative effects of vaccination that were significantly more favorable for the use of the antigen preparation PVM-I (p<0.05, Dunnett test for all analyses).

[0078] Example 6. Preparation and characterization of polypeptide variants based on isoforms 145, 165, 189 and 206 from VEGF-A. As shown in the alignment in Figure 1, all VEGF-A isoforms share the cysteine ​​residue present in isoform 121, suggesting that strategies used to increase immunogenicity may be effective across the entire protein family. To test this hypothesis, isoforms 145, 165, 189, and 206 were cloned from messenger ribonucleic acid (mRNA) of HeLa tumor cells. Automated sequencing verified the sequences according to those reported in the CCDS database (https: / / www.ncbi.nlm.nih.gov / CCDS / CcdsBrowse.cgi Release 22) following the June 2018 update (see December 23, 2019).

[0079] Briefly, the strategy described in Example 1 was used because all VEGF-A isoforms can be obtained in a single amplification reaction and are completely separable due to their different sizes. In all cases, the mutations described in Example 1 were introduced from the administration of the antigen preparation to eliminate the induction of VEGFR2 activation. The amino acid sequences corresponding to the proteins from which protein variants PVM145, PVM165, PVM189, and PVM206 were generated are designated SEQ ID NO:3, SEQ ID NO:4, SEQ ID NO:5, and SEQ ID NO:6, respectively.

[0080] After cloning the protein variants into the vector pM238 as described, two strategies were used to obtain the proteins: a) the method described by Morera et al. (Morera, et al., Angiogenesis, 2008:11:381-93) and b) the method described in Example 1. The resistance of the two purified variants to trypsin digestion was compared for each isoform. Analysis of the digests was performed by molecular exclusion chromatography. The results showed that when VEGF-A isoform variants were obtained according to the second purification strategy (PVM145-I, PVM165-I, PVM189-I, and PVM206-I), as described for PVM-I, resistance to trypsin digestion was observed at less than 5% of the total protein mass. This resistance was reflected by the presence of a signal corresponding to a "cysteine ​​knot" dimer, which is only obtained when disulfide bridges are established according to the naturally occurring configuration (signal at 23–25 min) (Table 9). Thus, when a protein preparation is obtained according to the method described in Example 1, the protein preparation is digested by incubation with trypsin by more than 95% compared to the counterpart produced by the original method described by Morera et al., as shown in Table 9. The absence of the canonical cysteine ​​knot peptides detailed in Table 3 (Example 1) was also confirmed by ESI-MS / S for trypsin digestion of antigen preparations PVM145-I, PVM165-I, PVM189-I, and PVM206-I, but was not detectable in preparations PVM145, PVM165, PVM189, and PVM206.

[0081] A comparative immunogenicity analysis of these variants in Freund's adjuvant was performed as described in Example 1. This study demonstrated superior neutralization of VEGF-A binding to VEGFR2 by sera from animals treated with variants less resistant to trypsin digestion (PVM145-I, PVM165-I, PVM189-I, PVM206-I) compared to their variants PVM145, PVM165, PVM189, and PVM206 (Table 9). [Table 9]

[0082] These proteins were used in BALB / c animals challenged with 10,000 CT26 cells by the subcutaneous route in an immunization and dosing schedule similar to that described in Example 5. The results showed that 28 days after tumor inoculation, the use of antigen variants together with the adjuvant sNAcGM3-VSSP induced superior antitumor effects in the case of variants PVM145-I, PVM165-I, PVM189-I, and PVM206-I compared to preparations PVM145, PVM165, PVM189, and PVM206 (Table 10). [Table 10]

[0083] Regardless of the VEGF-A isoform used, it was observed that variants that were not resistant to digestion with trypsin were the variants with the highest immunogenicity and antitumor efficacy.

[0084] Example 7. Preparation of PVM-I using oxidation and selection of trypsin-sensitive mutants. Due to the formation of multiple disulfide bonds, several conformations coexist in PVM-I protein preparations. These can be obtained and separated by a different method than that previously described in Example 1, involving controlled denaturation and renaturation steps in the presence of an oxidizing agent. For the smaller isoform of VEGF-A (SEQ ID NO: 2) expressed in E. coli, inclusion bodies were solubilized in 6 M guanidinium chloride, and the protein was purified by affinity for nickel in the presence of this denaturant using the manufacturer's recommendations (QIAGEN). The eluted product was then separated by reversed-phase chromatography (RP) on a preparative C18 column using a linear gradient between (A) trifluoroacetic acid (TFA) in water (0.088% v / v) and (B) 0.084% TFA in water, 90% acetonitrile (v / v). As expected, the fractions in which the protein of interest eluted showed no resistance to trypsin digestion, consistent with the denatured and reduced conditions of the protein preparation. The protein was then subjected to an oxidative regeneration process in Tris-HCl pH 8.6 in the presence or absence of guanidinium chloride and β-mercaptoethanol (0.2 mM) at 22°C for 20 hours. After changing the buffer to 10 mM Tris-HCl pH 7.4, disulfide bridge formation was assessed by size-exclusion chromatography on G25 Sepharose. The formation of "cysteine ​​knots" was analyzed by digestion with trypsin at a 50:1 ratio (protein:trypsin) in 200 mM Tris-HCl buffer, pH 8.0, for 16 hours, as described in Example 1. Thus, regeneration of the protein in the absence of guanidinium chloride exhibited 90% resistance to digestion (PVM-IA), whereas regeneration in the presence of denaturants resulted in different percentages of resistance, which decreased with increasing molar concentration of guanidinium chloride.

[0085] The most abundant fractions were those that were incompatible with trypsin digestion when 6 M guanidinium chloride was used. These fractions were then separated by RP-HPLC chromatography on a C18 preparative column. A total of 15 fractions were isolated and numbered according to their elution order in RP chromatography (F1–F15, Table 11).

[0086] After buffer exchange into 10 mM Tris pH 7.4, the PVM-IA and PVM-IO preparations, and fractions of the latter, were compared with PVM and PVM-I in a study using BALB / c mice. Protein administration was assessed using a conventional scheme of 200 μg of protein preparation in 100 μg of sNAcGM3-VSSP, with eight weekly doses. Table 11 shows the results of a comparative analysis of the titer of specific IgG against mouse VEGF-A, its inhibition of binding to VEGFR2, induction of cytotoxic cell responses in both lymph nodes and spleen, and the effect of immunization on subcutaneous CT26 tumor growth. [Table 11]

[0087] Table 11 demonstrates the superior efficacy of the PVM-I preparation compared with the PVM preparation, confirming the results shown in Example 1 in Freund's adjuvant. This study demonstrates the previously described immune effects as well as antitumor effects directly related to the induction of effective humoral and cellular responses. Analysis of the renaturalized protein preparation (PVM-IA) in the absence of reducing agents indicates that it is 7-10 times less immunogenic and less effective against tumors than PVM-I, and even PVM. However, use of the renaturalized preparation (PVM-IO) in the presence of 6 M guanidinium chloride yields results not significantly different from those described for PVM-I. Furthermore, evaluation of fractions of the PVM-IO preparation indicates that at least 12 of these (F4-F15) are equipotent in inducing specific humoral and cellular immune responses and generating antitumor responses.

[0088] Analysis of these fractions by trypsin digestion as described in Example 1 followed by Edman degradation (Marti, T., Rosselet, SJ, Titani, K., and Walsh, KA (1987) Biochemistry 26, 8099-8109) and mass spectrometry confirmed that in fractions F4 to F15, cysteines 2 and 4 of the standard structure of the cysteine ​​knot formed an intramolecular bridge, while in fractions F1 to F3, intermolecular bonds of C2-C2, C4-C4, C2-C4, C2-C9, and C4-C9 were detected.

[0089] The order of bonds in which cysteines participate in fractions F4 to F15 corresponds to the polypeptides identified in the sequence listing as SEQ ID NO: 18 to SEQ ID NO: 23, as detailed below. -F4 corresponds to SEQ ID NO: 18, in which the seventh cysteine ​​(position 161) of the two polypeptide chains, the eighth cysteine ​​(position 163) of the two polypeptide chains, and the last cysteine ​​(position 175) of the two polypeptide chains form an intermolecular disulfide bridge. -F5 corresponds to SEQ ID NO: 18, in which the seventh cysteine ​​(position 161) of one polypeptide chain and the eighth cysteine ​​(position 163) of another polypeptide chain and the last cysteines of the two polypeptide chains form an intermolecular disulfide bridge. -F6 corresponds to SEQ ID NO: 19, in which the seventh cysteine ​​(position 161) of the two polypeptide chains, the eighth cysteine ​​(position 163) of the two polypeptide chains, and the last cysteine ​​(position 175) of the two polypeptide chains form an intermolecular disulfide bridge. -F7 corresponds to SEQ ID NO: 19, in which the seventh cysteine ​​(position 161) of one polypeptide chain and the eighth cysteine ​​(position 163) of another polypeptide chain and the last cysteines of the two polypeptide chains form an intermolecular disulfide bridge. -F8 corresponds to SEQ ID NO: 20, in which the seventh cysteine ​​(position 161) of the two polypeptide chains, the eighth cysteine ​​(position 163) of the two polypeptide chains, and the last cysteine ​​(position 175) of the two polypeptide chains form an intermolecular disulfide bridge. -F9 corresponds to SEQ ID NO: 20, in which the seventh cysteine ​​(position 161) of one polypeptide chain and the eighth cysteine ​​(position 163) of another polypeptide chain and the last cysteines of the two polypeptide chains form an intermolecular disulfide bridge. -F10 corresponds to SEQ ID NO: 21, in which the seventh cysteine ​​(position 161) of the two polypeptide chains and the eighth cysteine ​​(position 163) of the two polypeptide chains form an intermolecular disulfide bridge. -F11 corresponds to SEQ ID NO: 21, in which the seventh cysteine ​​(position 161) of one polypeptide chain is linked to the eighth cysteine ​​(position 163) of another polypeptide chain to form an intermolecular disulfide bridge. -F12 corresponds to SEQ ID NO: 22, in which the seventh cysteines (position 161) of the two polypeptide chains and the eighth cysteines (position 163) of the two polypeptide chains are linked to form an intermolecular disulfide bridge. -F13 corresponds to SEQ ID NO: 22, in which the seventh cysteine ​​(position 161) of one polypeptide chain is linked to the eighth cysteine ​​(position 163) of another polypeptide chain to form an intermolecular disulfide bridge. -F14 corresponds to SEQ ID NO: 23, in which the seventh cysteines (position 161) of the two polypeptide chains and the eighth cysteines (position 163) of the two polypeptide chains are linked to form an intermolecular disulfide bridge. -F15 corresponds to SEQ ID NO: 23, in which the seventh cysteine ​​(position 161) of one polypeptide chain is linked to the eighth cysteine ​​(position 163) of another polypeptide chain to form an intermolecular disulfide bridge.

[0090] Example 8. Analysis of the relevance of cysteines 2 and 4 in generating immunogenic structures. The results described in Examples 1, 9, and 10 revealed that a common feature of the more immunogenic structures of PVM-I is that the intramolecular bonds of VEGF-A involve cysteines 2 and 4 (Cys110 and Cys119 of PVM-I) in the primary sequence. These residues appear to be frequently mutated in studies designed to obtain VEGF-A because they prevent dimeric aggregation of the protein (Jiang, et al., Biochemistry, 2010:49:6550-6; Wentink, et al., Proc Natl Acad Sci USA, 2016:113:12532-7). In the native conformation, these are the amino acids that form the interstrand bridges to generate VEGF-A dimers. Previous studies by Bequet-Romero et al. and Morera et al. only used primary sequence variants of VEGF-A containing these cysteines, but in neither of these was there any intrachain disulfide bond formation between cysteine ​​2 and cysteine ​​4 of the observed sequence. The presence of these amino acids in the sequence used to generate the antigen and the process by which they were obtained so that the mentioned cysteines form only intrachain bonds constitute a novel approach to obtaining proteins for vaccination purposes. To better characterize the relevance of these residues for the immunogenicity of the antigen preparation, we proceeded to obtain variants (by site-directed mutagenesis) containing the following cysteine ​​to alanine changes in the primary sequence: a) PVM-I Ala110 :Change of Cys110 by Ala110 b) PVM-I Ala119 :Change of Cys119 by Ala119 c) PVM-I Ala110、119 : Changes of Cys110 by Ala110 and Cys119 by Ala119

[0091] For this purpose, mutations were introduced by nested polymerase chain reaction (PCR) as described in (Bequet-Romero, et al., Angiogenesis, 2007:10:23-34) using the oligonucleotide pairs shown in Table 12. Briefly, PCRs 1 and 2 were performed using a plasmid containing the sequence encoding SEQ ID NO:2 as a template. PCR MasterMix (Qiagen) and a mixture of oligonucleotides shown in Table 12 were used. The amplified DNA in each reaction was purified after separation on an agarose gel from the reaction template. The resulting two DNA strands were used as templates for PCR 3 for each molecule listed in Table 12. 25 cycles of PCR were performed, and the resulting DNA was separated on an agarose gel and then digested with the enzymes NheI and BamHI (Promega) according to the manufacturer's instructions. The digested DNA was cloned into the vector pM238 as described (Morera, et al., Angiogenesis, 2008:11:381-93). The introduced mutations were verified by automated sequencing of the resulting DNA. The DNA encoding these mutant proteins was transformed into the BL21 strain (DE3), and polypeptides were produced according to the method described in Example 1 for PVM-I. The polypeptides corresponding to the sequences of SEQ ID NO: 15, SEQ ID NO: 16, and SEQ ID NO: 17 were used to produce the PVM-I preparation. Ala110 , PVM-I Ala119 and PVM-I Ala110、119 is obtained. [Table 12]

[0092] Preparations containing proteins with mutations at cysteines 110 and 119 were compared with PVM and PVM-I in a study in BALB / c mice, where administration was evaluated using a ratio of 200 μg of antigen preparation in 100 μg of the adjuvant sNAcGM3-VSSP, with eight administrations separated by one week. The comparative study was conducted in a subcutaneous CT26 colon cancer model with 10 animals per group. Table 13 shows the superior immunogenicity of the PVM-I preparation compared with PVM, which confirms the results shown in Example 1. Analysis of protein preparations with mutations at cysteines 110 and 119 showed that they were approximately two-fold less immunogenic and had lower antitumor effects than PVM-I and even PVM. [Table 13]

[0093] Example 9. Comparative evaluation of the preclinical use of antigen preparations PVM and PVM-I in the treatment of pulmonary metastases in mice. Metastasis is the fundamental cause of death from cancer of all origins. Among these, pulmonary metastasis is the disease that most often leads to patient death due to respiratory complications, tumor replacement of lung tissue, and accumulation of pleural fluid. Pulmonary metastasis constitutes one of the most resistant conditions for antiangiogenic interventions based on single administration of agents that remove VEGF-A receptor-transducing ligands or signal suppressors. In this context, incorporating the cellular branch of the immune response into the armamentarium of responses to vaccine immunization may play a crucial role.

[0094] To evaluate the efficacy of the vaccine preparations under analysis for the treatment of lung metastases in mice, groups of 12 mice were immunized with these preparations, and tumor challenge was performed according to the requirements of all models. The results of immunization with sNAcGM3-VSSP as an adjuvant are described below for the three metastasis models. Animals were immunized subcutaneously for 8 weeks with weekly doses containing 100 μg of the corresponding antigen and 100 μg of adjuvant in a total volume of 200 μL.

[0095] For studies in the C57B1 / 6 mouse strain, 250,000 cells of the 3LL metastatic lung carcinoma (clone D122) were inoculated into the footpad 3 days after the fourth immunization. The primary tumor was surgically removed 20 days after implantation, and the animals were sacrificed 15 days later for lung metastasis counting and characterization. Metastatic burden analysis was assessed by assessing lung weight and the number of macrometastases. Mitotic index studies were performed by measuring the number of mitoses per field in 10 fields per section per animal. Similarly, apoptosis was quantified per field, and the mitosis / apoptosis ratio was calculated.

[0096] Table 14 shows a comparison of lung weights between treated animals. Differential effects were observed between both antigens, favoring the antigen preparation PVM-I, which is enriched for a family of structural isomers that completely lack the disulfide bonds present in native VEGF-A. Histopathological characterization of metastatic lesions using hematoxylin-eosin staining and analysis of serial sections showed that the specific immune response to VEGF-A also led to a significant decrease in the mitotic / apoptotic balance observed in metastatic foci. [Table 14]

[0097] The establishment of experimental lung metastases was evaluated in BALB / c mice after intravenous inoculation of 20,000 CT26 cancer cells. Three days after the fourth immunization, antigen challenge was administered via the retroorbital plexus. Thirty days after tumor challenge, animals were sacrificed, and histopathological analysis of serial lung tissue sections assessed tumor burden in the lungs, as well as morphology, cellularity, angiogenesis, and the balance between mitosis and apoptosis. As shown in Table 15, in the analyzed sections (10 per animal), both the number and area occupied by metastases were reduced by immunization with PVM. The most significant effect was observed in mice immunized with the PVM-I preparation (p<0.05, Dunnett's post-hoc test). Interestingly, the mitotic index (number of mitoses per field) was very high in lung metastatic lesions induced by this model and was reduced approximately fourfold by administration of the vaccine preparation. Similar to the previous analysis, the best effect was observed in the group treated with the antigen variant PVM-I. [Table 15]

[0098] We also evaluated the F3II metastatic breast cancer model, which spontaneously metastasizes to the lungs, in BALB / c mice. This involved the use of a biweekly regimen with an additional group containing aluminum phosphate adjuvant. Three days after the fourth immunization (with sNAcGM3-VSSP) or the second immunization (with aluminum phosphate), a total of 200,000 cells were inoculated subcutaneously in 100 μL of DMEM culture medium.

[0099] Table 16 shows the analysis of primary tumor weights 28 days after tumor implantation. Treatment of animals with both antigen variants, as well as with both adjuvants and administration schemes, significantly reduces tumor growth, with the antigen preparation PVM-I showing a more significant inhibition (p<0.05, Dunnett's post-hoc test). The same table shows the results for the number of lung macrometastases. As can be seen, the reduction in tumor growth significantly reduces the number of implanted metastases. Furthermore, in these studies, a decrease in the mitotic / apoptotic balance was observed, and an increase in the number of necrotic foci associated with metastatic lesions was detected. All of these effects were more significantly detected in the group treated with the antigen preparation PVM-I. [Table 16]

[0100] Overall, the results showed that the antigen preparation PVM-I was more effective in reducing the metastatic potential of primary tumors and the implantation capacity of metastatic cells in the lungs. Furthermore, a superior effect of vaccination with the antigen preparation PVM-I was observed in altering the metastatic phenotype, with a reduced mitotic / apoptotic balance, a higher incidence of necrosis, and cellular foci with significantly reduced angiogenesis.

[0101] Example 10. Effect of immunization with antigen preparations PVM and PVM-I on tumor-induced immunosuppression. Groups of 15 animals of the BALB / c strain were used, which were immunized subcutaneously with the following mutants: 1. PVM in sNAcGM3-VSSP (8 immunizations, once a week) 2. PVM-I in sNAcGM3-VSSP (8 immunizations, once a week) 3. sNAcGM3-VSSP (8 immunizations, once a week) 4. PVM in aluminum phosphate (4 immunizations, once every 2 weeks) 5. PVM-I in aluminum phosphate (4 immunizations, once every 2 weeks) 6. Aluminum phosphate (four immunizations, once every two weeks)

[0102] In all cases, immunizations were performed subcutaneously in a total volume of 200 μL. Three days after the last immunization, five animals per group were randomly selected and euthanized to analyze their immunological status and that of experimental controls (groups 3 and 6). This evaluation was performed by flow cytometry on whole blood samples, spleen cells, lymph nodes, and bone marrow.

[0103] The remaining animals in each group received 20,000 CT26 colon cancer cells subcutaneously in the right flank. Five mice per group were euthanized and evaluated 14 and 21 days after tumor cell injection, with the additional analysis of dissociated tumor tissue.

[0104] No macroscopic toxicity events were evident in any of the animals, and histopathological analysis revealed no lesions in any of the organs analyzed 7 days after the last immunization. Immunological evaluation consisted of (1) analysis of serum concentrations of murine VEGF-A, (2) evaluation of T lymphocyte concentrations in whole blood, lymph nodes draining the primary tumor, and the tumor itself, and (3) study of IFN-γ secretion by intratumoral lymphocytes upon exposure to antigens representative of mutant VEGF-A.

[0105] In untreated animals, mouse VEGF-A levels increased with tumor exposure time, consistent with increased tumor size. In the group immunized with the antigen mutant, a significant decrease in VEGF-A levels (p<0.001 ANOVA, Dunnett's post-hoc test) was observed, which persisted for 30 days after tumor challenge (Table 17). A similar phenomenon was observed in the tumor compartment. [Table 17]

[0106] The immune status of the sacrificed animals at each time point was analyzed through studies of the proportion of cell populations present in the lymph nodes and the tumor itself, as reported by Gabrilovich et al. (Gabrilovich D et al. Blood 1998, 92:4150). In these studies, monoclonal antibodies against CD3, CD4, and CD8 labeled with fluorescein isothiocyanate, phycoerythrin, and phycoerythrin-Cy7 were used to visualize the cell populations using a flow cytometer (Sysmex Partec). The results obtained are shown in Table 18. [Table 18]

[0107] Analysis of lymphoid cell populations in animals 14 days after tumor challenge showed an increase in CD3-, CD4-, and CD8-positive cell compartments in the three analyzed compartments, directly related to administration of the vaccine preparation. The antigen variant PVM-I significantly increased the concentration of these immune system cells (p<0.05 Bonferroni post-test). Furthermore, study of the PD1 marker in CD8 T lymphocytes infiltrating subcutaneous tumors showed that these increases were associated with a negative PD1 phenotype in a higher percentage of antigen-treated animals receiving the PVM-I antigen (Table 18). The increase in CD4 and CD8 cell infiltrates was significantly and positively associated with a decrease in tumor burden (r=0.7147, p<0.002).

[0108] Tumor-derived leukocytes were isolated by a column-based positive selection process using magnetic beads coated with a CD45-specific antibody (Miltenyi), and IFN-γ secretion was evaluated in response to the addition of mutated VEGF-A in the culture medium using an ELISPOT-type system (MABTECH). Increased secretion of this cytokine was observed only in animals vaccinated with the antigen, and was higher for both adjuvants in the group administered with the PVM-I antigen (p<0.05, Bonferroni post-test) (Table 19). This indicated not only increased cellular infiltration, but also that these infiltrating leukocytes had higher activity in response to the administration of the vaccine preparation. [Table 19]

[0109] Example 11. Evaluation of the adjuvant effect of immunization with PVM-I on the immunogenicity generated against another antigen challenge. Considering the possibility of immune recovery, we investigated whether immunization with PVM-I could induce an adjuvant effect involving increased humoral and cellular responses specific not only to VEGF-A but also to unrelated antigens. This potential effect was tested by assessing the degree of specific immunity against antigens present in the tumor and those co-administered with the vaccine.

[0110] The ovarian tumor model ID8-OVA was used for tumor challenge of C57B1 / 6 animals and the parental strain ID8 in an in vitro lysis assay. Animals (n=10 / group) received an intraperitoneal tumor challenge of 2.5 million cells, as described below, and the immunization schedule began 3 days later. a) PVM-I (200 μg), sNAcGM3-VSSP (100 μg): subcutaneously administered once a week for 8 weeks b) OVA (1 mg), sNAcGM3-VSSP (100 μg): subcutaneously administered once a week for 2 weeks. sNAcGM3-VSSP (100 μg): subcutaneously administered once a week for the remaining 6 weeks. c) PVM-I (200 μg), sNAcGM3-VSSP (100 μg), OVA (1 mg): administered subcutaneously once a week for 2 weeks, and PVM-I (200 μg), sNAcGM3-VSSP (100 μg): administered subcutaneously once a week for the remaining 6 weeks. d) sNAcGM3-VSSP (100 μg): subcutaneously administered once a week for 8 weeks e) OVA (1 mg): subcutaneously administered weekly for 2 weeks. sNAcGM3-VSSP (100 μg): subcutaneously administered once weekly for the remaining 6 weeks.

[0111] One week after the eighth immunization and 60 days after tumor implantation, the animals were euthanized. In this tumor model, animal weight correlates with tumor growth. Table 20 shows the observed antitumor effect, which is coordinately increased by administering OVA in the presence of PVM-I in sNAcGM3-VSSP.

[0112] Analysis of humoral responses for VEGF-A and OVA was performed using an ELISA-type system in collected serum as described in Examples 1, 2, and 3. Specific cellular responses were analyzed by in vitro co-incubation of splenocytes with ID8 and ID8-OVA tumor cell lines using specific cytolytic test tumors (described in Example 4). Additionally, after 72 hours of specific stimulation (peptide OVA), 257-264 or VEG FKDR- The secretion of IFN-γ in the supernatants of splenocytes isolated from immunized animals was examined in response to the addition of IFN-γ. In the latter case, an ELISA-type assay was used for the detection of IFN-γ according to the manufacturer's instructions (Biolegend). The results are summarized in Tables 20 and 21. [Table 20] [Table 21]

[0113] Humoral (Table 20) and cellular (Table 21) responses were induced against both antigens when administered in the context of sNAcGM3-VSSP. Coadministration of OVA with immunotherapy containing PVM-I resulted in a significant increase in the specific response to OVA not only in animals receiving OVA as the immunogen, but also in animals receiving only PVM-I. This fact indicates the potential of a vaccine strategy using this antigen preparation to induce an amplified response to antigens not found in the vaccine preparation but expressed by tumors. The humoral responses, as well as IFN-γ secretion and ID8-OVA cytolysis in groups treated with both antigens, indicate that the combination not only has an additive effect, but also cooperates to induce a superior effect. This factor indicates the potential use of this strategy in increasing immune responses to other tumor-associated antigens.

[0114] Example 12. In vivo protection experiments in a collagen-induced arthritis model by immunization with PVM or PVM-I in two adjuvants. Groups of 20 DBA / 1 mice (H-2q haplotype), which are susceptible to collagen-induced arthritis, were immunized. The animals were administered the test antigen preparation (PVM or PVM-I) adjuvanted in aluminum phosphate or sNAcGM3-VSSP. For sNAcGM3-VSSP, a scheme of 100 μg of antigen and 100 μg of adjuvant was used, administered eight times at weekly intervals. For aluminum phosphate, 0.7 mg equivalent of Al was administered. 3+ The mice were given 100 μg of antigen in a dose of 100 μg twice a week for four doses. Treatment groups were defined as follows: I. PVM of sNAcGM3-VSSP injection II. PVM-I of sNAcGM3-VSSP injection III.sNAcGM3-VSSP IV. PVM in Aluminum Phosphate V. PVM-I in aluminum phosphate VI. Aluminum phosphate

[0115] Three days after the fourth immunization with NAcGM3-VSSP or the second immunization with aluminum phosphate, induction of autoimmune arthritis was initiated by immunization with chicken collagen II (Sigma) according to a previously described model (Campbell IK et al. Eur. J. Immunol. 30:1568, 2000). This immunization was repeated on day 26 to complete the induction of the autoimmune response. The four paws of the mice were evaluated daily for the presence of signs of erythema (1), inflammation (2), or joint stiffness (3) on examination according to an arthritis index that established a score of 0 to 3 for each paw, with a maximum of 12. Mice began to show clinical signs of arthritis development 23 days after induction, reaching a maximum incidence at day 50. Table 22 shows an analysis of the incidence of arthritis in animals from different experimental groups. On days 40 and 55, a significant reduction in the incidence of arthritis was observed in the vaccinated groups (I, II, IV, and V) compared with the control groups III and VI, which received placebo in the corresponding adjuvant. [Table 22]

[0116] Example 13. Evaluation of the immunogenicity of PVM and PVM-I in non-human primates and analysis of the effect of immunization on experimental laser-induced choroidal neovascularization. The ability of the antigen preparations PVM and PVM-I to induce relevant immune responses was evaluated in a more autologous setting in non-human primates (Chlorocebus aethiops sabaeus). Four animals were used per group. 400 μg of antigen preparation PVM or PVM-I was mixed with 200 μg of NAcGM3-VSSP or 0.7 mg of aluminum phosphate in the form of Al. 3+Equivalent doses were administered. The administration schedule was once weekly for 8 doses for NAcGM3-VSSP and once every 2 weeks for aluminum phosphate for 4 doses. Specific IgG antibody titers against VEGF-A and neutralizing binding to type 2 receptors were assessed at week 8 post-immunization (NAcGM3-VSSP) or week 4 post-immunization (aluminum phosphate). The results are shown in Table 23. As can be seen, both antigen preparations induced specific titers against VEGF-A, with those induced by PVM-I being higher in both adjuvants. These titers similarly correlated with increased serum neutralizing capacity. [Table 23]

[0117] To assess the extent to which activation of cellular responses was achieved in parallel with induction of humoral responses, conventional delayed type cutaneous reaction (DTH) tests were performed as described (Morera, et al., Vaccine, 2010:28:3453-61). At 48 hours, two perpendicular measurements of induration were taken using a digital Vernier caliper. The area of ​​the lesions was calculated and their geometric mean was reported. Erythema or inflammation was not considered part of the reaction. A diameter of 0.5 mm was considered the limit of a detectable reaction. Results were classified according to the following criteria: (++) > 5 mm 2 , (+)=0.5~4.99mm 2 The results are shown in Table 24. 2 Punches (6 per animal) were made and analyzed by hematoxylin / eosin staining. At least two sections from each biopsy were analyzed to determine the nature of the infiltrate with respect to the presence of mononuclear cells, neutrophils, or eosinophils.

[0118] Intradermal injection of the antigen was well tolerated, with no adverse events such as blisters or ulcers. All groups of monkeys treated with the vaccine preparation, in different combinations with adjuvants, responded to human VEGF-A. No reactions were reported in any of the immunized groups at the site of saline injection, as a control. Histopathological evaluation confirmed the presence of a strong DTH-type response to VEGF-A inoculation. Biopsies obtained from the site of human VEGF-A injection were consistent with a DTH scene, with abundant infiltration of macrophages and lymphocytes. [Table 24]

[0119] Peripheral blood mononuclear cells (PBMCs) from immunized animals were isolated by Ficoll gradient as previously reported, pre-incubated with VM, and analyzed for specific cytolysis of syngeneic cells marked with CFSE. Table 25 shows the results corresponding to the analysis of the survival of CFSE-labeled "charged" cells by flow cytometry. Analysis of at least three replicates per animal showed that the vaccine preparation induced specific cellular immunity against the antigen with cytotoxicity within the week after the last immunization, corresponding to the induction period. As can be seen, in both adjuvants, the vaccine effect was significantly greater when the PVM-I antigen was used (P<0.05, Dunnett). [Table 25]

[0120] These animals were also used to evaluate the prevention of experimental laser-induced choroidal neovascularization in nonhuman primates. The model described by Krzystolik et al. (Krzystolik MG, et al. 2006. Acta Ophthalmol, 120:338-346) was used. Four monkeys were used per experimental group. For all procedures, animals were anesthetized with intramuscular injections of ketamine hydrochloride, acepromazine maleate, and atropine sulfate. Topical proparacaine hydrochloride anesthesia was also used. One week after the final immunization, choroidal neovascularization (CNV) was induced in the macula by argon laser burn. Photography and fluorescein angiography were used to detect and measure the extent of the lesions and to evaluate their characteristics. The development of CNV lesions was evaluated before and after laser treatment and at 15, 20, and 29 days thereafter. Lesions were analyzed by an expert outside the experimental design using the following scale: grade 1 = no hyperfluorescence, grade 2 = hyperfluorescence without exudation, grade 3 = early or mid-pass hyperfluorescence and late leakage, and grade 4 = very bright early or mid-pass hyperfluorescence with late leakage extending beyond the edge of the laser spot. Animals were observed daily to assess their clinical condition, including any ocular abnormalities.

[0121] Of the four grades assigned to laser treatment, grade 4 corresponds to clinically significant leakage. The lesions likely reflect the presence of new choroidal vessels that have either grown beyond the laser-treated area or are leaking so severely that fluorescein significantly extends beyond the vessels. The mean number of grade 4 lesions in the placebo group ranged from 45.4% to 50.2% of the laser-treated area. The mean percentage of grade 4 lesions in the control group was similar to that reported by other authors using this animal model of CNV. In contrast, all groups treated with the vaccine preparations, regardless of the adjuvant used, showed a significant reduction or complete disappearance of grade 4 lesions. Table 26 shows the distribution percentages of all lesion grades at day 29 for the treatment groups. Interestingly, the use of vaccine preparation PVM-I resulted in a significantly greater reduction in higher grade lesions (ANOVA, Bonferroni post-hoc test, p<0.05). [Table 26]

[0122] Example 14. Treatment of solid tumors in humans by administration of a vaccine composed of the antigen preparation PVM-I and an adjuvant. This study aimed to evaluate the potential of a new antigen preparation to induce relevant specific immune responses in the context of tumor pathology in humans. Patients with advanced solid tumors, lacking other treatment options, and a treatment-free period of at least 4 weeks prior to treatment initiation were selected. They received either (a) a weekly inoculation scheme for 8 weeks, followed by monthly dosing in sNAcGM3-VSSP, or (b) a four-dose schedule of once every 2 weeks, followed by monthly dosing, at a dose of 800 μg of the antigen preparation PVM-I in aluminum phosphate.

[0123] Humoral and cellular immune responses were assessed in serum and PBMCs collected both before the start of the immunization schedule and one week after the final immunization. Two basic parameters were assessed in serum: specific antibody titers against VEGF-A and the ability to neutralize binding to the VEGFR2 receptor. These parameters were determined using two ELISA-type systems (Morera, et al., Vaccine, 2012:30:368-77; Sanchez Ramirez, et al., J Immunoassay Immunochem, 2016:37:636-58). A titer of greater than 1:500 and a neutralization percentage greater than 10% were defined as positive results.

[0124] Cellular responses were assessed in PBMC samples by IFN-γ ELISPOT as described (Gavilondo, et al., Vaccine, 2014:32:2241-50). Low responders were defined as patients with 29-39 IFN-γ-secreting clones per million CD3+ cells, intermediate responders as patients with 40-79 signals, and high responders as patients with >80 signals.

[0125] Administration of PVM-I in both adjuvants resulted in the establishment of a specific immune response against VEGF-A, as measured by specific antibodies (Figure 4A) and their ability to neutralize the interaction of the growth factor with VEGFR2 (Figure 4B). This response was observed in 70% of patients treated with either adjuvant. Similarly, a significant increase in the responsiveness of PBMCs to stimulation with a mutant form of human VEGF-A (VM) was observed (Figure 4C). These results demonstrate that the use of the novel antigen mutant PVM-I in humans induces a specific immune response against human VEGF-A.

[0126] Figure 5 summarizes the total number of patients and how many of them tested positive for any of the tests for each adjuvant. The results were stratified taking into account the number of tests with significantly higher results when using pre-immunization values ​​as a control. In the graph, each patient is also assigned the number of months they survived after the start of administration of the vaccine preparation (analysis at 24 months). This allows us to correlate the extent to which the specific immune response contributes to increased survival in the absence of other tumor-specific treatments. A significant increase in survival is observed in patients with a higher number of positive responses (p<0.05, Kaplan-Meier).

[0127] Example 15. Clinical use of antigen preparation PVM-I in the treatment of age-related macular degeneration in humans This study aimed to evaluate the potential of the antigen preparation PVM-I to induce relevant specific immune responses in the setting of age-related macular degeneration (AMD) in humans. Patients received 125 μg of bevacizumab in 50 μL via the intravitreal route once monthly for three consecutive months. Administration of this drug or other antiangiogenic drugs, such as aflibercept, ranibizumab, or ramucirumab, during the first three months after diagnosis is established as conventional treatment for the initial control of this disease. After these treatments, patients were systematically monitored for visual acuity and retinal structural phenotypic changes, and depending on these changes, further intravitreal injections of antiangiogenic drugs were considered appropriate. Treatment success was largely determined by a reduction in the need for intravitreal administration.

[0128] This treatment was administered at a dose of 400 μg, either simultaneously with or without the antigen preparation PVM-I. (Group 2) sNAcGM3-VSSP was administered weekly for 8 weeks, followed by monthly re-administration, or (Group 3) aluminum phosphate was administered every 2 weeks for 4 weeks, followed by monthly re-administration. Group 1 received only conventional treatment, consisting of monthly administration for 3 consecutive months. After the first 3 doses of bevacizumab, patients were re-administered bevacizumab only if their lesions worsened compared with the previous ophthalmological analysis. This deterioration was defined as decreased visual acuity, increased retinal thickness, the presence of intraretinal fluid, or increased subretinal fluid (greater than 200 μm or the value at the previous examination).

[0129] Administration of the antigen preparation with the indicated adjuvants and regimen significantly reduced the number of intravitreal injections of bevacizumab required to maintain or improve the quality of patient vision and the integrity of the macula. Table 27 summarizes the number of bevacizumab injections required per patient after the third bevacizumab administration and the specific antibody titers against VEGF-A that induce immunity. The use of immunotherapy with antigen preparation PVM-I significantly reduced the number of intravitreal injections of bevacizumab administered to patients compared with the conventional scheme (Group 1) (p<0.05 in both cases, Dunnett's test). In both cases, a significant correlation was observed between the reduction in the number of intravitreal injections and the level of specific antibodies to human VEGF-A in the patient's serum, starting one week after the fourth immunization in group 2 (Pearson r = -0.8788, p = 0.0008) and after the eighth immunization in group 3 (Pearson r = -0.7894, p = 0.0066). [Table 27]

[0130] Approximately 30% of the patients included in the study had active extrafoveal polyps, a factor that worsens the prognosis for recovery in patients who responded to conventional intravitreal therapy because polypoidal lesions originate in the deeper layers of the retina. Interestingly, in an analysis of patients with choroidal polypoid neovascularization, the number of active lesions significantly decreased one year after treatment initiation; this effect was observed in only 20% (1 in 5) of patients in the conventionally treated group (Table 28). [Table 28] [Sequence List Free Text]

[0131] Sequence Listing 1 <223> Description of artificial sequences: nucleic acids encoding variant polypeptides Sequence Listing 2 <223> Description of artificial sequences: mutant polypeptides Sequence Listing 3 <223> Description of artificial sequence: mutant polypeptide isoform 145 Sequence Listing 4 <223> Description of artificial sequence: mutant polypeptide isoform 165 Sequence Listing 5 <223> Description of an artificial sequence: Mutant polypeptide isoform 189 Sequence Listing 6 <223> Description of artificial sequence: mutant polypeptide isoform 206 Sequence Listing 7 <223> Description of artificial sequences: oligonucleotides Sequence Listing 8 <223> Description of artificial sequences: oligonucleotides Sequence Listing 9 <223> Description of artificial sequences: oligonucleotides Sequence Listing 10 <223> Description of artificial sequences: oligonucleotides Sequence Listing 11 <223> Description of artificial sequences: oligonucleotides Sequence Listing 12 <223> Description of artificial sequences: oligonucleotides Sequence Listing 13 <223> Description of artificial sequences: oligonucleotides Sequence Listing 14 <223> Description of artificial sequences: oligonucleotides Sequence Listing 15 <223> Description of the artificial sequence: A mutant polypeptide in which Cys110 is changed to Ala Sequence Listing 16 <223> Description of the artificial sequence: A mutant polypeptide in which Cys119 is changed to Ala Sequence Listing 17 <223> Description of the artificial sequence: A mutant polypeptide in which Cys110 and 119 are changed to Ala Sequence Listing 18 <223> Writing artificial sequences: rearrangements of mutant polypeptides 1 Sequence Listing 18 <223> Cys161, Cys163, and Cys175 are part of the intermolecular bond Sequence Listing 19 <223> Writing artificial sequences: rearrangements of mutant polypeptides 2 Sequence Listing 19 <223> Cys161, Cys163, and Cys175 are part of the intermolecular bond Sequence Listing 20 <223> Writing artificial sequences: rearrangements of mutant polypeptides 3 Sequence Listing 20 <223> Cys161, Cys163, and Cys175 are part of the intermolecular bond Sequence Listing 21 <223> Writing artificial sequences: rearrangements of mutant polypeptides 4 Sequence Listing 21 <223> Cys161 and Cys163 are part of the intermolecular bond Sequence Listing 22 <223> Writing artificial sequences: rearrangements of mutant polypeptides 5 Sequence Listing 22 <223> Cys161 and Cys163 are part of the intermolecular bond Sequence Listing 23 <223> Writing artificial sequences: rearrangements of mutant polypeptides 6 Sequence Listing 23 <223> Cys161 and Cys163 are part of the intermolecular bond

Claims

1. 1. A polypeptide comprising a functional variant of an isoform of human vascular endothelial growth factor A (VEGF-A), wherein amino acids corresponding to Arg 82, Lys 84, and His 86 of a wild-type VEGF-A molecule are substituted with Glu, thereby rendering said human vascular endothelial growth factor A unable to bind to VEGFR2; A polypeptide that folds in a non-native arrangement of disulfide bridges in which the second and fourth cysteines of the mutant polypeptide chain are found only as part of an intramolecular bridge and the seventh and eighth cysteines of the mutant are found only forming an intermolecular bond.

2. The human VEGF-A isoform is VEGF-A １２１ , VEGF-A １４５ , VEGF-A １６５ , VEGF-A １８９ and VEGF-A ２０６ 2. The polypeptide of claim 1, wherein the polypeptide is selected from the group consisting of:

3. 3. The polypeptide of claim 2, characterized in that it comprises a rearrangement of intramolecular disulfide bridges in which the first cysteine ​​is linked to the second cysteine, the third cysteine ​​is linked to the fourth cysteine, the fifth cysteine ​​is linked to the sixth cysteine, and the seventh cysteines of two polypeptide chains, the eighth cysteines of two polypeptide chains, and the last cysteines of two polypeptide chains are linked to form intermolecular disulfide bridges.

4. 3. The polypeptide of claim 2, characterized in that it comprises a rearrangement of intramolecular disulfide bridges in which the first cysteine ​​is linked to the second cysteine, the third cysteine ​​is linked to the fourth cysteine, the fifth cysteine ​​is linked to the sixth cysteine, and the seventh cysteine ​​of one polypeptide chain is linked to the eighth cysteine ​​of another polypeptide chain, linking the last cysteines of the two polypeptide chains to form an intermolecular disulfide bridge.

5. 3. The polypeptide of claim 2, characterized in that it comprises a rearrangement of intramolecular disulfide bridges in which the first cysteine ​​is linked to the second cysteine, the third cysteine ​​is linked to the fifth cysteine, and the fourth cysteine ​​is linked to the sixth cysteine, as well as the linking of the seventh cysteines of two polypeptide chains, the eighth cysteines of two polypeptide chains, and the last cysteines of two polypeptide chains to form intermolecular disulfide bridges.

6. 3. The polypeptide of claim 2, characterized in that it comprises a rearrangement of intramolecular disulfide bridges in which the first cysteine ​​is linked to the second cysteine, the third cysteine ​​is linked to the fifth cysteine, the fourth cysteine ​​is linked to the sixth cysteine, and the seventh cysteine ​​of one polypeptide chain is linked to the eighth cysteine ​​of another polypeptide chain, linking the last cysteines of the two polypeptide chains to form an intermolecular disulfide bridge.

7. 3. The polypeptide of claim 2, characterized in that it comprises a rearrangement of intramolecular disulfide bridges in which the first cysteine ​​is linked to the second cysteine, the third cysteine ​​is linked to the sixth cysteine, and the fourth cysteine ​​is linked to the fifth cysteine, as well as the linking of the seventh cysteines of two polypeptide chains, the eighth cysteines of two polypeptide chains, and the last cysteines of two polypeptide chains to form intermolecular disulfide bridges.

8. 3. The polypeptide of claim 2, characterized in that it comprises a rearrangement of intramolecular disulfide bridges in which the first cysteine ​​is linked to the second cysteine, the third cysteine ​​is linked to the sixth cysteine, the fourth cysteine ​​is linked to the fifth cysteine, and the seventh cysteine ​​of one polypeptide chain is linked to the eighth cysteine ​​of another polypeptide chain, linking the last cysteines of the two polypeptide chains to form an intermolecular disulfide bridge.

9. 3. The polypeptide of claim 2, characterized in that it comprises a rearrangement of intramolecular disulfide bridges in which the first cysteine ​​is linked to the last cysteine, the second cysteine ​​is linked to the third cysteine, the fourth cysteine ​​is linked to the fifth cysteine, and the seventh cysteines of two polypeptide chains and the eighth cysteines of two polypeptide chains are linked to form an intermolecular disulfide bridge.

10. 3. The polypeptide of claim 2, characterized in that it comprises a rearrangement of intramolecular disulfide bridges in which the first cysteine ​​is linked to the last cysteine, the second cysteine ​​is linked to the third cysteine, the fourth cysteine ​​is linked to the fifth cysteine, and the seventh cysteine ​​of one polypeptide chain is linked to the eighth cysteine ​​of another polypeptide chain to form an intermolecular disulfide bridge.

11. 3. The polypeptide of claim 2, characterized in that it comprises a rearrangement of intramolecular disulfide bridges in which the first cysteine ​​is linked to the last cysteine, the second cysteine ​​is linked to the fourth cysteine, and the third cysteine ​​is linked to the fifth cysteine, and the seventh cysteines of two polypeptide chains and the eighth cysteines of two polypeptide chains are linked to form an intermolecular disulfide bridge.

12. 3. The polypeptide of claim 2, characterized in that it comprises a rearrangement of intramolecular disulfide bridges in which the first cysteine ​​is linked to the last cysteine, the second cysteine ​​is linked to the fourth cysteine, the third cysteine ​​is linked to the fifth cysteine, and the seventh cysteine ​​of one polypeptide chain is linked to the eighth cysteine ​​of another polypeptide chain to form an intermolecular disulfide bridge.

13. 3. The polypeptide of claim 2, characterized in that it comprises a rearrangement of intramolecular disulfide bridges in which the first cysteine ​​is linked to the last cysteine, the second cysteine ​​is linked to the fifth cysteine, and the third cysteine ​​is linked to the fourth cysteine, and the seventh cysteines of two polypeptide chains and the eighth cysteines of two polypeptide chains are linked to form an intermolecular disulfide bridge.

14. 3. The polypeptide of claim 2, characterized in that it comprises a rearrangement of intramolecular disulfide bridges in which the first cysteine ​​is linked to the last cysteine, the second cysteine ​​is linked to the fifth cysteine, the third cysteine ​​is linked to the fourth cysteine, and the seventh cysteine ​​of one polypeptide chain is linked to the eighth cysteine ​​of another polypeptide chain to form an intermolecular disulfide bridge.

15. 15. The polypeptide of any one of claims 3 to 14, comprising an amino-terminal segment to increase its expression in bacteria and a carboxy-terminal segment to facilitate purification.

16. 16. The polypeptide of claim 15, wherein the amino terminal segment for increasing its expression in bacteria has the amino acid sequence of SEQ ID NO:

24.

17. 17. An antigen preparation comprising at least one polypeptide according to any one of claims 3 to 16 and a pharmaceutically acceptable excipient or diluent.

18. 18. A pharmaceutical composition comprising the antigen preparation of claim 17 and a pharmaceutically acceptable vaccine adjuvant.

19. 19. The pharmaceutical composition of claim 18, wherein the vaccine adjuvant is selected from the group consisting of an oil adjuvant, an aluminum salt, a proteoliposome, and a proteoliposome conjugated to a ganglioside.

20. 18. An antigen preparation according to claim 17 for treating diseases whose progression is associated with increased angiogenesis, inflammation and immunosuppression or for restoring the immune system.

21. Use of an antigen preparation comprising at least one polypeptide according to any one of claims 3 to 16 for the manufacture of a medicament for the treatment of a disease whose progression is associated with increased angiogenesis, inflammation and immunosuppression.

22. 22. The use according to claim 21, wherein the disease whose progression is associated with increased angiogenesis, inflammation and immunosuppression is selected from the group consisting of cancer, macular degeneration, diabetes, rheumatoid arthritis and edema.

23. The antigen preparation described in claim 20, wherein the disease whose progression is associated with increased angiogenesis, inflammation and immunosuppression is selected from the group consisting of cancer, macular degeneration, diabetes, rheumatoid arthritis and edema.

24. 20. Use of an antigen preparation comprising at least one polypeptide according to any one of claims 3 to 16 for the manufacture of a medicament for the restoration of the immune system.

Images