WO2023077206A1 - Genetically modified epitopes and multiepitope proteins, immunogenic compositions and use thereof - Google Patents

Abstract

This invention consists of genetically modified epitopes and multiepitope proteins composed of these genetically modified epitopes to avoid the formation of neoepitopes. The multiAAA protein is composed of genetically modified epitopes by addition of three alanine residues. The multiGPGPG protein is composed of genetically modified epitopes by addition of glycine-proline-glycine-proline-glycine residues. Both proteins are formulated as immunogenic compositions. This invention aims at optimising the vaccinal efficacy and immunological potency of a vaccine against leishmaniasis through the use of its more potent and promiscuous epitopes, having a wider world population coverage, and further aims at solving the problem of the absence of a vaccine against human visceral leishmaniasis, a lethal disease that is spreading in Brazil and the world.

Description

GENETICALLY MODIFIED EPITOPES AND MULTIEPITOPE PROTEINS, IMMUNOGENIC COMPOSITIONS AND THEIR USE

Summary of the Invention

[001] The present invention deals with the development of recombinant multiepitope vaccines, DNA or synthetic, formed through the junction of the most promiscuous T cell epitopes of the Nucleoside hydrolase protein of Leishmania (L.) donovani (NH36), characterized by being recognized by at least 70% of human HLA class II histocompatibility molecules, but which also bind to human HLA class I molecules, with potential use as recombinant DNA or synthetic vaccines against human leishmaniasis and animal antigenic markers for diagnosis and cure of leishmaniasis

[002] These epitopes are genetically modified sequences of the NH36 protein by adding amino acid sequences composed of Alanine-Alanine-Alanine (AAA) or Glycine-Proline-Glycine-Proline-Glycine (GPGPG). In this way, the epitopes are separated by spacer amino acid sequences, to avoid the formation of junctional epitopes or neoepitopes. Mutiepitope proteins have potential use in vaccination and immunotherapy of leishmaniasis, immunodiagnosis of leishmaniasis, control of cure and control of transmission blockade. The invention also deals with the applications of NH36 multiepitope vaccines in cross-protection against other microorganisms that have in their Nucleoside hydrolases, common sequences with Nucleoside hydrolase of Leishmania donovani (NH36).

Description of the State of the Art

[003] Visceral leishmaniasis (VL) is a neglected tropical disease caused by protozoa belonging to the Leishmania (L.) donovani complex, has high lethality rates when untreated and represents a serious public health problem worldwide, as according to ALVAR et al . (2012) ranks second in mortality and fourth in morbidity among neglected tropical diseases. [004] In Brazil, VL is caused by the parasite Leishmania (L.) infantum chagasi, which is transmitted to humans through the bite of the female sand fly Lutzomiya longipalpis. Leishmania is an intracellular parasite that infects cells of the mononuclear phagocytic system, especially macrophages.

[005] The initial interaction between the parasite and host cells is one of the factors that determine whether the individual will develop a Th1-type protective immune response characterized by the production of pro-inflammatory cytokines: IL-12, IFN-y, TNF- α, or susceptibility to the disease will develop, where the predominance of anti-inflammatory cytokines IL-4, IL-10, TGF-[3 (BHATTACHARYA et al., 2016).

[006] Most infected individuals do not develop the disease. They are asymptomatic or subclinical (CHAKRAVARTY et al., 2019) and have a positive delayed hypersensitivity skin test (DTH+). In individuals with clinical VL, the disease is characterized by fever, anemia, leukopenia, thrombocytopenia, hepatosplenomegaly (SUNDAR & SINGH, 2018) and a strong suppression of the cellular immune response (ALVAR et al., 2012) with negative DTH (DTH -). Epistaxis, gingival bleeding, petechiae, tachycardia, dry cough and, less frequently, abdominal discomfort may also occur (GRIENSVEN & DIRO, 2012). The disease is severe and has high mortality rates if not treated, however, after treatment, individuals convert DTH to positive and do not become infected again and only a minority experience relapse (BURZA et al., 2018).

[007] The epidemiological control of the disease is based on the treatment of patients, vector control and elimination of infected dogs. Thus, the development of a vaccine to prevent the disease, as well as the development of immunotherapies, is necessary in order to improve prevention strategies.

[008] Leishmaniasis comprises a set of neglected tropical diseases caused by parasites of the genus Leishmania, and cause different clinical manifestations ranging from cutaneous lesions such as: Diffuse Cutaneous Leishmaniasis (DCL), Disseminated Cutaneous Leishmaniasis (DLC), Mucocutaneous leishmaniasis (ML) or the systemic and more severe form of the disease, Visceral leishmaniasis (VL)( BURZA, CROFT, & BOELAERT, 2018; THAKUR et al, 2018; LESTINOVA et al, 2017; LOCKARD et al, 2019) .

[009] The pathogens, hosts and vectors involved in the vectorial transmission cycle of leishmaniasis are associated with the natural ecosystem (PURSE et al, 2017) and will determine the development of the different clinical forms of leishmaniasis, as well as the different immune responses developed by the host .

Biological and Epidemiological Aspects of Visceral Leishmaniasis

[010] VL is an infectious parasitic disease of chronic evolution that can be fatal when untreated (JAIN & JAIN, 2017), being caused by an obligate intracellular parasite belonging to the order Kinetoplastida, family Trypanosomatidae, genus Leishmania, subgenus Leishmania ( MURRAY et al., 2005; READY, 2013; GUIMARÃES-E-SILVA et al., 2017). The Leishmania species involved in disease transmission belong to the donovani complex, which includes: Leishmania (L.) donovani occurring in the Old World (Asia and Africa) and Leishmania (L.) infantum in the Old World (Mediterranean Basin, Middle East, Central Asia) and Leishmania (L.) infantum chagasi (South America and Central America) (THAKUR et al, 2018; BURZA et al., 2018).

[01 1] VL is undergoing an expansion and urbanization process due to the presence of disorganized occupation in peripheral areas with native vegetation, (CASARIL et al, 2014), climate changes and socioeconomic factors (DA SILVA et al, 2017 ), 3), widespread deforestation (PURSE et al, 2017), migration of infected dogs and humans from endemic areas to areas where the disease has not yet been reported (SALOMON et al, 2015), and migration of non-immune people in areas with existing transmission cycles (WHO, 2020). [012] Another important factor associated with the expansion of VL is the comorbidity with other infectious diseases (TOEPP et al, 2019), mainly with HIV/AIDS, which has been reported in different studies conducted in Ethiopia (DIRO et al., 2014; DIRO et al., 2019) and Iran (REZAEI et al., 2018). Additionally, in Brazil there has been an increase in the number of cases of VL/HIV co-infection (SOUSA-GOMES et al., 2017).

[013] VL is endemic in tropical and subtropical countries in Africa, Asia, and Latin America. It is estimated that between 200 and 400 thousand new cases of the disease occur per year (KARAGIANNIS-VOULES et al., 2013) and that 95% of these are distributed in only 10 countries: Bangladesh, Brazil, China, Ethiopia, India, Kenya, Nepal, Somalia, South Sudan and Sudan (WHO, 2020).

[014] In Latin America, Brazil is responsible for more than 90% of annual cases of VL (BURZA et al, 2018; WHO, 2020) and cases of the disease have already been registered in 26 states of the federation and in the Federal District (KARAGIANNIS - VOULES et al., 2013). The Northeast region concentrates the highest number of cases of the disease, followed by the Southeast and Midwest regions (SINAN, 2017).

[015] Worldwide, the implementation of strategies for the prevention and control of VL remains insufficient, resulting in the maintenance of a high mortality rate and geographic expansion of the disease (SANTOS et al, 2019). As a consequence, the prophylaxis strategies used are mainly restricted to the control of potential reservoirs (ANDRADE & TEIXEIRA, 2012), use of collars with pyrethroids in dogs, spraying of insecticide to reduce the population of vectors (ROCK et al, 2015), the use of insect protection screens, as well as the use of curtains and repellents (WHO, 2020).

[016] VL transmission is caused by vectors belonging to the order: Diptera; family: Psychodidae; genus Phlebotomus, and genus Lutzomyia described in the New World (Lestinova et al, 2017; Sumova et al, 2018; BURZA et al, 2018). In Brazil, the main vector of the VL pathogen is the sand fly Lutzomyia longipalpis, followed by Lutzomyia cruzi (MISSAWA et al, 201 1 ). [017] In Brazil, the etiologic agent of VL is Leishmania (L.) infantum chagasi, whose cycle alternates between an extracellular flagellated promastigote form that grows in the sandfly vector, and the intracellular amastigote form that develops in dogs and foxes (ANDRADE & TEIXEIRA, 2012; RODRIGUES et al, 2016; LOCKARD et al, 2019; DAYAKAR ET AL, 2019).

Immunobiology of Visceral Leishmaniasis

[018] In experimental VL, the saliva of L. longipalpis induces a significant attraction of macrophages mediated by a high expression of chemokines such as chemokine CC ligand 2 (CCL2) and monocyte chemotactic protein-1 (MCP-1) (LESTINOVA et al , 2017), which help recruit cells from the host's mononuclear phagocytic system, including neutrophils, which internalize the parasites in an attempt to fight the infection (SHARMA et al, 2016). However, the continuous replication of amastigotes within macrophages leads to apoptotic cell death of the host cell (RODRIGUES et al., 2016).

[019] Macrophages can also phagocytize through the recognition of proteins present in the Leishmania membrane, such as lipophosphoglycan (LPG) and glycoprotein gp63 (SHIO et al., 2012), and bind to receptors of the complement system (CR1 , CR3 and C3b) (SARAH et al., 2004). The increase in the parasite load of infected mice is also correlated with the increase in pattern recognition receptors, for example, the toll-like receptors TLR2, TLR4 and TLR9 (CEZÁRIO et al., 2011 ), present in the cytoplasm and membrane of macrophages .

[020] But in an attempt to escape from the immune system, Leishmania blocks the maturation of the complement system by attacking the formation of the complex (C5-C9) and modifies the signaling of toll-like receptors (TLR-2/TLR4), altering the expression of the profile of chemokines and cytokines (DAYAKAR ET AL, 2019), favoring parasite survival. [021] However, when Leishmania phagocytosis occurs, macrophages and dendritic cells, major antigen presenting cells (APC), interact with naive T lymphocytes via major histocompatibility complex class II (MHCII) and T cell receptor (TCR), in addition to co-stimulatory molecules such as B7.1/B7.2-CD28 and CD40-CD40L. From these interactions, naive T cells undergo differentiation into helper T cells (T helper-Th) and produce cytokines that will determine a profile of protective immune response (Th1 ) or susceptibility (Th2) to VL (BHATT ACHARYA et al., 2016).

[022] The Th1-type protective immune response is associated with the production of pro-inflammatory cytokines TNF-α, IL-1 |3, IL-6 and IFN-y, induces the activation of microbicidal mechanisms of macrophages, such as the release of oxygen radicals, (DAYAKAR et al 2019), and increased production of nitric oxide (RODRIGUES et al., 2016), culminating in the death of the parasite.

[023] Recent studies have shown that another population of T cells help protect against VL, the Th17 cells, which have a pleiotropic nature, often associated with the recruitment of neutrophils (DAYAKAR et al., 2019) and which may contribute to both aspects, protective and damage (KUMAR AND NYLÉN, 2012). However, when there is a failure of the protective immune response of the Th1 type , there is an increase in susceptibility to the disease with the predominance of a response of the Th2 profile, which is characterized by the development of a humoral immune response, with an increase in the production of antibodies and of the prevalence of anti-inflammatory cytokines such as: IL-4, IL-10 and TGF-[3 (DAYAKAR ET AL, 2019).

[024] The cytokine IL-10, on the other hand, is directly related to the process of pathogenesis of VL, as it reduces the expression of the MHCII receptor and the co-stimulatory molecules B7 and CD40 present on the surface of macrophages, making them not responsive to activation signals. Additionally, IL-10 inhibits the functions of the pro-inflammatory cytokines TNF-α and IFN-y that control the expression of inducible nitric oxide synthase (iNOS), thus decreasing NO production (DAYAKAR ET AL, 2019), which facilitates parasite replication and disease progression (KUMAR & NYLÉN, 2012). [025] Although the process of activation and differentiation of CD4 ⁺ T cells of the Th1 profile, which act in the control of Leishmania infection, occurs in parallel with the activation of naive CD8 ⁺ T cells by dendritic cells (DCs). DCs secrete the type I cytokines IL-12 and IFN and induce the differentiation of CD8 ⁺ T lymphocytes into cells that further contribute to the production of IFN-Y, TNF-α, in addition to secreting perforin and granzyme, which kill the cells. infected cells (RODRIGUES et al., 2016).

[026] Additionally, activated CD4 ⁺ and CD8 ⁺ T cells produce IFN-y, which in combination with IL-1 [3, induces the production of IL-27 by macrophages, inhibiting the generation of Th17 cells and promoting together with to IL-21 the generation of IL-10-secreting T cells (KUMAR AND NYLÉN, 2012).

Clinical Manifestations of Visceral Leishmaniasis

[027] VL has different clinical forms depending on the immune response and the signs and symptoms developed by the infected individual, which can be: asymptomatic, oligosymptomatic or subclinical infection and the progressive disease, which is classic VL. The asymptomatic form of the disease comprises most individuals who are infected by L. (L.) donovani or L. (L.) infantum and never develop the disease, suggesting that genetic factors are involved in resistance and susceptibility (KUMAR AND NYLÉN, 2012). These individuals are positive for the Leishmanin skin test (LST), which is a delayed-type hypersensitivity test (DTH+) (ROCK et al, 2015; CHAPMAN et al., 2018), also known as the Montenegro test (BURZA et al., 2018). al., 2018).

[028] The remainder of infected individuals develop a progressive chronic infection (MEDINA-COLORADO et al., 2017), which is characterized by a failure of the cellular immune response to Leishmania antigens and an elevated production of IL-10 in the mononuclear cells of Leishmania. peripheral blood (PBMC) and plasma, promoting the development of a Th2 profile response (NYLÉN et al., 2007). [029] The disease can have an acute or insidious onset and the incubation period ranges from 2 weeks to 8 months (BURZA et al, 2018). The classic clinical signs and symptoms of VL include persistent fever, loss of appetite, cachexia, enlargement of the spleen (splenomegaly) and liver (hepatomegaly), pallor, edema (MEDINA-COLORADO et al., 2017; JAIN & JAIN, 2017; PRESTES-CARNEIRO et al.2019; SANTOS et al, 2019), in addition to other alterations such as tachycardia, petechiae, epistaxis, gingival bleeding and less frequently abdominal discomfort (GRIENSVEN & DIRO, 2012) and diarrhea, which occurs as a result of intestinal parasitism and ulceration (KUMAR & NYLÉN, 2012).

[030] In the advanced stage of the disease, thrombocytopenia together with prothrombin depletion leads to severe mucosal hemorrhage, jaundice and ascites (RODRIGUES et al., 2016). In addition, severe anemia and loss of leukocytes make individuals immunosuppressed, facilitating the development of bacterial infections that are a common cause of death in lethal cases of VL (KUMAR & NYLÉN, 2012; BURZA et al, 2018).

[031] Laboratory findings include thrombocytopenia, pancytopenia, anemia, thrombocytopenia, and neutropenia (BURZA et al, 2018; ZIJLSTRA, 2016) that are frequent and reflect both splenic sequestration and suppression of bone marrow function, in addition to polyclonal hypergammaglobulinemia that is associated with the presence of abundant plasma cells in the spleen (RODRIGUES et al., 2016).

Treatment of Visceral Leishmaniasis

[032] The drugs most commonly used in the treatment of VL are the pentavalent antimonials available in formulations: sodium stibogluconate and meglumine antimoniate (BURZA et al., 2018) and amphotericin B, which has been considered the two first-choice drugs for decades. Paramomycin and miltefosine are also used (VAKILI et al., 2017). [033] Pentavalent antimonials (SbV) such as sodium stibogluconate, have high toxicity associated with life-threatening conditions, such as cardiac arrhythmias, prolongation of the QT interval (QTc), ventricular premature beats, ventricular tachycardia, fibrillation (SUNDAR & SINGH, 2018 ) but also, its prolonged use increased the number of cases of resistance to the antimonial. In Bihar, for example, 64% of cases are resistant to antimonials (MOHAPATRA, 2014; TASLIMI et al, 2018; PURKAIT et al, 2015).

[034] On the other hand, conventional amphotericin B was introduced for the treatment of VL in the 1980s (MONGE-MAILLO & LOPEZ-VÉLEZ, 2013) and more recently its liposomal formulation AmBisome (Gilead Sciences, USA) was developed, which proved have much better therapeutic efficacy. However, liposomal amphotericin B is still very expensive (MANSURI et al., 2017) and presents, like conventional amphotericin B, a toxicity profile that includes infusion reactions in most patients, nephrotoxicity, hypokalemia, myocarditis and occasional death.

[035] This high toxicity and risk require strict monitoring (SUNDAR & SINGH, 2018). Currently, liposomal amphotericin B is recommended by the WHO as the first-line drug to treat any form of VL, not only in severe or HIV-associated conditions, but also in immunocompetent children and adults.

[036] Recently, multidrug therapy has been explored for the treatment of VL from the use of drugs with synergistic or additive activity, with the aim of reducing the time and cost of treatment, which also reduces the occurrence of adverse effects (SUNDAR & CHAKRAVARTY, 2015) and increases the treatment efficacy rate even in cases of co-infection (SINGH et al, 2016). However, these drugs have high cost, high toxicity (JAIN & JAIN, 2017) and selection of resistant mutants (SINGH et al., 2016; SERENO et al, 2019). [037] Most parasites of the Protozoa sub-kingdom are deficient in the “de novo” synthesis of purine systems, and therefore depend exclusively on the salvage pathway (BERENS et al; 1995). The Nucleoside Hydrolase of L. (L.) donovani (NH36), is a DNA metabolism enzyme that has 314 amino acids (CUI et al., 2001 , SANTANA et al., 2002) and presents significant homology with the sequences of other organisms such as: L. (L.) major (95%) (CUI et al., 2001 ), L. (L.) chagasi (99%), L (L.) infantum (99%), L (L.) .) tropica (97%), L (L.) mexicana (93%), L (L.) braziliensis (84%) (BLAST, 2010), T. brucei (27%) and Crithidia fasciculata (80%) ( CUI et al., 2001). Still having 68% identity with Haemophylus influenzae and 30% identity with Bacillus anthracis (TODD et al., 2003; BLAST, 2010). Nucleoside-Hydrolases hydrolyze the A/-glycosidic bond of purines and pyrimidines, resulting in ribose and the base (GOTTLIEB, 1989).

[038] The activity of Nucleoside hydrolases had already been identified in Trypanosoma brucei gambiense cells (SCHMIDT et al., 1975), L. (L.) donovani (KOSZALKA and KRENITSKY 1979; LOOKER et al., 1983), L ( L) mexican (DAVIES et al., 1983), L tropica (GARIN et al., 2001 ), Trypanosoma cruzi (MILLER et al., 1984) and Trypanosoma brucei brucei (PARKIN, 1996). The demonstration of the constitutive presence of Nucleoside hydrolase in promastigotes of L. (L.) donovani (CUI et al., 2001 ; SANTANA et al., 2002) indicates its relevance for the development of the parasite, since it is expressed in the stages of the parasite that infect the vertebrate host (LUKES et al., 2007; MAURÍCIO et al., 2006; SHI et al., 1999; CUI et al., 2001; SANTANA et al., 2002).

[039] The synthetic nucleoside analogues called Imucillin A and Imucillin H demonstrated a strong inhibitory activity of Nucleoside Hydrolase (NH36) in vitro (Freitas et al., 2015a) and a strong inhibition of the growth of promastigotes of Leishmania (L.) donovani and L. (L.) amazonensis in vitro, and a high therapeutic efficacy against VL caused by L.(L) infantum chagasi in vivo, in hamsters and mice with absence of toxicity and against cutaneous leishmaniasis by L. (L. ) amazonensis in mice (da Silva F., 2018), making them excellent candidates for new therapeutic drugs against leishmaniasis. Vaccines against Leishmaniasis

[040] Research on vaccines against leishmaniasis began many years ago. The first vaccine against leishmaniasis was developed by Professor Adler of the Hebrew University of Jerusalem, Israel (reviewed by PALATNIK-de-SOUSA et al. 2008) from live Leishmania to cause an injury, in the hope of generating protection. However, this practice was discontinued because it was not considered safe. There is currently only one vaccine against visceral leishmaniasis licensed in Uzbekistan (reviewed by PALATNIK-de-SOUSA et al., 2008).

[041] The first generation of leishmaniasis vaccines was developed against tegumentary leishmaniasis (LT) and was obtained by manipulating dead parasites with or without adjuvants. These vaccines were used in clinical trials in human populations in endemic areas against the cutaneous form of the disease mainly (ANTUNES et al., 1986; CONVIT et al., 1987; VÉLEZ et al., 2000), but also against the visceral form ( KHALIL et al., 2000).

[042] The second generation of anti-Leishmania vaccines can be divided into three categories according to their composition: live vaccines containing Leishmania genes or using genetically modified Leishmania or viruses expressing Leishmania genes (CRUZ, COBURN, BEVERLEY, 1991; SOUZA et al., 1994; TITUS et al., 1995; RAMIRO et al., 2003; SELVAPANDIYAN et al., 2004; SELVAPANDIYAN et al. 2006; JAIN & JAIN, 2015), vaccines of defined recombinant or synthetic fractions or subunits ( RUSSO et al., 1991; SKEIKY et al.,1995; GURUNATHAN et al.„ 1997; COLER et al., 2002; RAFATI et al., 2002; GRADONI et al., 2005; BADARÓ et al., 2006; POOT et al., 2006; COLER et al., 2007; MORENO et al., 2007; JAIN & JAIN, 2015) and vaccines with partially purified native fractions (PALATNIK et al., 1989; JAFFE, RACHAMIM, SARFSTEIN, 1990; JARDIM et al., 1991; LEMESRE et al., 2005; PALATNIK-de-SOUSA et al., 2008; PALATNIK-de-SOUSA et al., 2009; PALATNIK-de-SOUSA, et al., 2009; JAIN, & JAIN, 2015). [043] Finally, third-generation vaccines contain antigen genes cloned into a eukaryotic promoter vector injected and translated directly into the muscle (XU & LIEW, 1994; HANDMAN, 2001; RAMIRO et al., 2003; AGUILAR-BE et al., 2005; RAFATI et al., 2005; GAMBOA-LÉON et al., 2006; PALATNIK-de-SOUSA, 2012; JAIN, JAIN, 2015). DNA vaccines offer a simple and effective means of inducing long-term immunity and providing protection, in addition to being a reliable, safe, stable and easily produced form of antigen-specific immunotherapy (KUMAR & SAMANT, 2016).

[044] Previously, a DNA vaccine based on the NH36 gene was developed that was immunotherapeutic against canine VL (BORJA CABRERA et al., 2012) and which was the basis of a patent granted in Mexico (Dumonteil E. et al. , Instituto Mexicano de la Propriedad Industrial (IMPI). YU/a/2004/000010, Folio:YU/E/2004/000078).

[045] Even with so many advances in studies based on vaccine formulations, there is still no vaccine available against human VL or American LT, used in clinical routine. Only a first-generation vaccine was licensed for human TL immunotherapy in Brazil, in 2001, and it demonstrated its effectiveness in preventing the disease (MAYRINK et al., 2013).

[046] More recently we see examples of different types of vaccines developed against VL, vaccines can be composed of live attenuated Leishmania (AVISHEK et al., 2016), subunit and RNA vaccines (DUTHIE et al., 2018), DNA vaccines (KUMAR & SAMANT, 2016; KAUR et al., 2016), synthetic peptide vaccines (De BRITO et al., 2018), synthetic epitope vaccines (CAMPOS et al., 2018; PALATNIK-DE-SOUSA, 2019; JOSHI et al., 2019).

Canine VL vaccines

[047] The Laboratory of Biology and Biochemistry of Leishmania at the Institute of Microbiology Paulo de Góes at UFRJ has been developing second-generation vaccines against murine VL and canine VL (LVC) using the FML antigen (Ligante de Fucose and Mannose) (PI1 100173-9), glycoprotein GP36 and recombinant protein NH36 from Leishmania (L) donovani (P 1015788-3) (PALATNIK-de-SOUSA et al., 2008).

[048] The FML antigen of L. (L.) donovani is a glycoprotein complex, which inhibits the infection of macrophages (PALATNIK et al., 1989; PALATNIK-de-SOUSA, DUTRA, BOROJEVIC, 1993) and is sensitive and specific in the diagnosis of VL (PALATNIK-de-SOUSA et al., 1995) and CVL (CABRERA et al., 1999).

[049] A veterinary vaccine composed of the FML antigen and the acylated and deacylated QS21 saponins of Quillaja saponaria Molina (OLIVEIRA-FREITAS et al., 2006), was licensed in Brazil in 2003, under the name of Leishmune®, as the first vaccine registered for the prevention of CVL in the world (PI1 100173-9), (reviewed by PALATNIK-de-SOUSA et al., 2008). It is well tolerated (PARRA et al., 2007) and is also a transmission blocker (PI 0503187-7, Brazil, MX/a/2007/014375, EP1893640 B1 (EPO), PT1893640E (Portugal), ES2540088T3.0 (Spain) ), EPIT1893640 (Italy), EP1893640 (France).

[050] EP1893640 is currently deposited in national phases in the United Kingdom, Germany, Greece, Bulgaria, Turkey (SARAIVA et al., 2006; PALATNIK-de-SOUSA et al., 2008). Leishmune® is a transmission blocking vaccine because the antibodies it induces in healthy dogs block the development of the parasite in the gut of the sandfly vector and thus the insect does not pass on the infection to other humans or dogs, and the epidemic is interrupted in endemic areas (SARAIVA et al., 2006; PALATNIK-de-SOUSA et al., 2008; it is well tolerated (PARRA et al., 2007) and also blocks transmission (PI 0503187-7, Brazil, MX/a /2007/014375, EP1893640 B1 (EPO), PT1893640E, ES2540088T3.0, EPIT1893640, EP1893640).

[051] This prophylactic vaccine demonstrated in the field vaccine efficacy of 76% (da SILVA et al., 2000) and 80% (BORJA-CABRERA et al., 2002). The trials were carried out in areas of high infectious pressure and the calculation of vaccine efficacy was based on the number of deaths and confirmed clinical cases of the disease (da SILVA et al., 2000; BORJA-CABRERA et al., 2002; PALATNIK-de- SOUSA et al., 2008). [052] Vaccination with Leishmune® leaves dogs non-infectious for the insect (NOGUEIRA et al., 2005; PALATNIK-de-SOUSA et al., 2008) and when administered with double concentration of adjuvant it is immunotherapeutic (BORJA-CABRERA et al., 2004, SANTOS et al., 2007). When administered together with Glucantime and Allopurinol, Leishmune® is negative for the polymerase chain reaction (PCR) for Leishmania DNA, generating sterile protection (BR 1 1 2012 000872- 2, US 10,172,940 B2) (BORJA-CABRERA et al., 2010).

[053] Very important evidence of the effectiveness of Leishmune® is the demonstration that in endemic areas where Leishmune® was applied, a significant reduction in canine cases, canine seroprevalence and, consequently, human cases and human mortality from VL was detected (SARAIVA et al., 2006; PALATNIK-de-SOUSA et al., 2008) confirming that the vaccine has the capacity to interrupt the epidemic.

[054] A second prophylactic vaccine licensed in Brazil in 2008, Leish-Tec®, contains recombinant protein A2 from amastigotes of L. (L.) donovani (PI 0603490-0 A2) and has been shown to be protective against VL in trials with experimental challenges in mice (GHOSH, LABRECQUE, MATLASHEWSKI, 2001 ; COELHO et al., 2003), rhesus monkeys (GRIMALDI et al., 2014) and Beagle dogs (FERNANDES et al., 2008). The first field trial of Leishtec® was published only in 2016 and was carried out in an area of low infectious pressure (REGINA-SILVA et al., 2016).

[055] In this area, VL cases were defined, not by deaths or clinical symptoms with confirmation of parasites as was used for the Leishmune® vaccine (da SILVA et al., 2000; BORJA-CABRERA et al., 2002; PALATNIK -de-SOUSA et al., 2008), but by seropositivity confirmed by at least one of three parasitological tests, or by the addition of xenodiagnosis to parasitic assays (REGINA-SILVA et al., 2016). [056] Using these less severe criteria, the authors reported vaccine efficacy of 71 .4% based on the parasitological definition of cases, and 58.1% by adding xenodiagnosis. However, when a field test of Leishtec® was performed in an area of high infectious pressure, a smaller effect was observed in reducing the rate of canine infection (GRIMALDI et al., 2017). In fact, the incidence of VL was 42% in controls and 27% in vaccinated dogs, representing only 35.7% of vaccine efficacy (GRIMALDI et al., 2017).

[057] Protective responses were either "not generated" or "not maintained" in 43% of immunized dogs that became infected and developed the disease over time. However, A2 proved to be immunogenic since the levels of specific anti-A2 IgG antibodies were significantly higher in vaccine recipients than in infected control dogs (GRIMALDI et al., 2017).

[058] Therefore, in contrast to Leishmune®, which showed efficacy of 76% to 80%, based on strict final parameters (deaths and clinical cases) (da SILVA et al., 2000, BORJA-CABRERA et al., 2002) and promoted a decrease in the canine and human incidence of VL in areas of high infectious pressure (PALATNIK-de-SOUSA et al., 2009), the second Leishtec® field trial showed that this vaccine may not reduce the canine incidence of leishmaniasis in areas of high transmission and may not have an impact on the human incidence of the disease (GRIMALDI et al., 2017).

[059] A recent study on hunting dog owners infected with L(L) infantum showed that, if used for immunotherapy, Leishtec ® reduced the risk of progression to clinical symptoms by 25%, and mortality by 70% (TOEPP et al., 2018). In contrast, a higher potency was described for Leishmune® which, when used for immunotherapy, reduced 80% in symptomatic cases, 100% in positive cases for parasites and 100% in deaths and, when used in combination with chemotherapy, also promoted a sterile cure with negative PCR results (BORJA-CABRERA et al., 2010). [060] This evidence explains why many comparative studies have revealed results of greater efficacy for Leishmune® vaccines than for Leishtec® and / or LBsap vaccines (FERNANDES et a!., 2014; de MENDONÇA et al., 2016), in addition to a stronger induction of the immune and pro-inflammatory response (MOREIRA et al., 2016; COSTA PEREIRA et al., 2015, de LIMA et al., 2010) and improvement of the proportions of CD8 + T cells that secrete IFN- y (ARAÚJO et al., 2009).

[061] In addition, a recombinant protein, called Q, formed by the genetic fusion of five intracellular antigenic fragments of the acidic ribosomal proteins Lip, Lip2b, P0 and histone H2A (CARCELÉN et al., 2009; FERNÁNDEZ-COTRINA et al., 2018) was licensed as the Letifend® vaccine with an efficacy of 72% (FERNÁNDEZ COTRINA et al., 2018) (LU93259C).

[062] Finally, the CaniLeish® vaccine, composed of the excreted proteins of L. (L.) infantum (ESP) and the purified extract of Quillaja saponaria (QA-21 ), also showed a dominant Th1 immune response that persisted for one year integer (GRADONI, 2015) (W02002100430A1 ). A field trial conducted in highly endemic areas of the Mediterranean basin used PCR and parasite cultures to confirm infection (OLIVA et al., 2014). Considering active infection, an efficacy of 63% of the vaccine was observed at month 24 after vaccination.

[063] Considering the symptoms of VL, the efficacy of the vaccine was 68.4% (RODRIGUES et al., 2016). According to these results, it was recognized that only two vaccines, consisting of fractions purified by the parasite with saponin-derived adjuvants, were shown to confer significant protection against disease and death under natural conditions, the FML-saponin (Leishmune®) in Brazil, and the LiESP / QA -21 (CaniLeish®) in Europe (GRADONI, 2015).

Vaccines containing GP36 and Leishmania (L.) donovani Nucleoside hydrolase (NH36) antigens

[064] In parallel with the development of the Leishmune ^(R) vaccine, we described that the main antigen of the FML complex is the 36 kDa glycoprotein of L. (L.) donovani, which is recognized by human and canine VL sera (PALATNIK-de - SOUSA et al., 1996; SANTANA et al., 2002; NICO et al., 2014b) and was cloned in Escherichia coli, revealing to be a Nucleoside hydrolase (NH36), serological marker of CVL (SANTANA et al., 2002).

[065] When cloned in the VR1012 vector (AGUILAR-BE et al., 2005) it was presented as a DNA vaccine, which, like the FML and NH36 vaccines formulated with saponin, protects against murine infection by L. (L.) infantum chagasi, L. (L.) mexicana (AGUILAR-BE et al., 2005) and L. (L.) amazonensis (SOUZA, PALATNIK-de-SOUSA, 2009; NICO et al., 2014a; NICO et al. , 2014b). The FML and VR1012 NH36 vaccines therefore generate bivalent prevention against VL and LT (AGUILARBE et al., 2005; SOUZA; PALATNIK-de-SOUSA, 2009) and immunotherapeutic cure against murine VL (GAMBOA-LEÓN et al ., 2006) and LVC (BORJA-CABRERA et al., 2009, 2012).

[066] The NH36 of L. (L.) donovani is composed of 314 amino acids (CUI, et al., 2001 ) and demonstrates high homology with the sequences described for the NH of L. (L) major (95%) ( CUI et al., 2001 ), L. (L.) chagasi (99 %), L. (L.) infantum (99 %), L. (L.) tropica (97 %), L. (L.) mexicana ( 93 %) and L. (V.) braziliensis (84 %) BLAST (2014).

[067] In order to identify the NH36 epitopes to be included in a synthetic vaccine, two combined methodologies were used: 1) the cloning of immunogenic fragments and 2) the prediction of epitopes by programs whose algorithm detects sectors with the highest frequency of hydrophobic and charged amino acids.

[068] Initially, three fragments of the NH36 sequence were cloned into the pET28b vector for expression of histidine-tagged proteins: F1 (amino acids 1 to 103); the F2 (aa 104 to 198) and the F3 (aa 199 to 304). The three-dimensional model revealed that, in the NH36 tetramer, the F3 domain has the largest exposed area (29,507,002 Angstroms ² ). It is followed by the F1 domain (N-terminal) with an area of 27132781 Angstroms ² . [069] The F2 domain (central) has the smallest surface area (19931451 Angstroms ² ) and is the least exposed fragment of the protein. The F3 sequence includes three predicted epitopes for mouse CD4 ⁺ T cells and three epitopes for antibodies, whereas the F1 shows two epitopes for CD4+, one for mouse CD8+, and two for antibodies (NICO et al., 2014b, PI 1015788- 3).

[070] Vaccination of BALB/c mice with F1, F2 and F3 fragments and saponin demonstrated that protection against L. (L.) infantum chagasi is related to its C-terminal domain (F3) and is mainly mediated by a CD4+ T cell response with a smaller contribution from CD8+ T cells (NICO et al., 2010). Immunization with F3 optimized and exceeded by 36.73% the protective response induced by the cognate protein NH36. Increases in IgM, IgG2a, IgG 1 and IgG2b antibodies, proportions of CD4+ T cells, IFN-y secretion, ratios of IFN-y/IL-10 producing CD4+ and CD8+ T cells and inhibition percentages were detected. of antibody binding by predicted synthetic epitopes (NICO et al., 2010).

[071] The increase in intradermal staining (IDR) and in the ratios of CD4+ T cells producing TNF-α/IL-10 were the strong correlates of protection, which was confirmed by the in vivo depletion assay with monoclonal antibodies, by the epitopes for CD4+ and CD8+ predicted by programs based on Sette's algorithm and by the pronounced decrease in parasitic load (90.5-88.23 %). This protection was long lasting (NICO et al. 2010, PI 1015788-3).

[072] Due to the high sequence homology of the NHs, the protection generated against VL by L. (L.) infantum chagasi could be extended to other species that cause LT. As confirmed by the in vivo depletion assay with anti-CD4+ and anti-CD8+ antibodies, we observed that the protection generated by NH36 against murine infection by L. (L.) amazonensis is mediated by the combination of the CD4+ T response against the F3 domain and the response TCD8+ against the F1 domain (NICO et al., 2010, 2014a, PI 1015788-3). Therefore, the F3 domain of NH36 is required for protection against VL and the F3 and F1 domains for prevention against murine TL. [073] The genome of L. (L.) amazonensis was later elucidated and the amino acid sequence of NH36 showed 93% identity with that of NH A34480 of L. (L.) amazonensis (NICO et al., 2014a). There are completely conserved epitopes in it for CD4+ and CD8+ T cells in the F1 domain, and for CD4+ T cells, showing differences in only one amino acid, in the F1 and F3 domains. This homology justifies the strong cross-immunoprotective effect developed by the F1 and F3 fragments in the prevention of murine LT. Additionally, the F1 and F3 peptides also determined a strong cross-effect in the immunotherapy of L. (L.) amazonensis infection (NICO et al., 2014b).

[074] Additionally, dogs infected with L(L) infantum chagasie treated by immunotherapy with the NH36 DNA vaccine developed significant increases in the size and proportions of DTH-+- reactions and in the frequencies of NH36-specific CD4+ T cells in response to the F3 domain , and increases in IgG2/IgG1 ratios of anti-NH36 antibodies and CD8+ T cell counts in response to the F1 domain. Immunotherapy promoted a decrease in parasitic load and weight loss, increasing the survival of dogs (BORJA-CABRERA et al., 2010 b).

[075] The F1 and F3 domains were additionally expressed in tandem in an optimized recombinant chimera cloned in the pET28b vector, which was the most potent formulation in the vaccination against L. (L.) infantum chagasi, in the generation of anti-NH36 lgG2a antibodies , in the IDR to the Leishmania lysate, in the secretion of IFN-y, in the induction of the lowest levels of TNF-α with the stimulus of the F3-3 subpeptide (C-terminal portion of the F3 peptide) and of IL-10, and in the most potent reduction in the number of parasites in the liver, sharing its predominance with F3 in reducing hepatosplenomegaly. The chimera overcame the protective effect generated by vaccination with each peptide individually, or by the addition of both, suggesting that presentation in tandem and with optimized codons greatly enhances vaccine efficacy against murine VL (GOMES, 2013). [076] The chimera also promoted greater protection than that generated by the two domains mixed or by each domain independently, against challenge by L. (L.) amazonensis showing the highest production of anti-NH36 IgA, IgG and lgG2a, IDR, IFN-Y and IL-10, while TNF-α was more secreted by mice vaccinated with F3 or vaccines containing F3 (ALVES-SILVA et al., 2017). Furthermore, the chimera and the F1 vaccine induced the highest proportions of CD4+ and CD8+ T cells monoproducing IL-2, TNF-α or IFN-y; double producers of TNF-α and IL-2, TNFa and IFN-y or IFN-y and IL-2, and CD4+ multifunctional T cells triple producers of IL-2, TNF-α and IFN-y. Chimera and mixed domains also reduced the size of skin lesions (84% and 80%, respectively) and parasite burden (99.8%). Thus, the chimera also optimized cross-protection against L. (L.) amazonensis (ALVES-SILVA et al., 2017).

[077] But while protection against L. (L.) infantum chagasi and L. (L.) amazonensis infection requires a Th1 response, it is L. (V.) braziliensis infection that triggers an increased inflammatory response with strong secretion of IFN-y and TNF-α (ALVES-SILVA et al., 2019). Furthermore, the production of IL-10, which modulates the immune response, in L. (V.) braziliensis infection is reduced.

[078] Confirming the beneficial effect of vaccination with the F1 F3 chimera, an increase in IgA, IgG and IgG3 responses and IDR, 49% stronger than NH36, was observed. However, whereas Th1 responses were stronger with high rates of I FN-y/l L-10 and TNF-α/IL-10 in the case of the F3 and F1 vaccines, the F1 F3 chimera promoted the greatest IL secretion -10, which reduced the pathological Th1 response and characterized the induction of a mixed regulatory and/or T-cell response (ALVES-SILVA et al., 2019).

[079] We then identified the epitopes responsible for these immune responses in mice. The F3 vaccine induced immunity earlier and after the challenge, but the F1 F3 chimera promoted the highest cytokine-secreting CD4+ and CD8+ T cell responses, and the predominant frequencies of CD4+ and CD8+ IL-2+ T cells and multifunctional IL- 2+TNF-α+IFN-Y+. [080] Also as observed against L. (L.) amazonensis infection, the F1 F3 chimera promoted the greatest reduction in the size of L. (V.) braziliensis-induced ear lesions. The results confirmed the potential use of the F1 F3 chimera in a multispecies cross-protective vaccine against VL and LT (ALVES-SILVA et al., 2019).

Recombinant human vaccine under development that contains NH36

[081] The decrease in the proportions of total CD4+ T lymphocytes and Leishmania-specific T CD4+ lymphocytes, characteristic of VL, indicate that an effective vaccine should reinforce the T CD4+ immune response (KUMAR & NYLÉN, 2012). In this scenario, it is important to note that, recently, a fusion protein composed of NH36 from L. (L.) donovani and a sterol 24-c-methyltransferase (SMT), adjuvanted by GLA-SE (glucopyranosyl lipid A stabilized in oil- water), and called Leish-F3 (NH36-SMT-GLA-SE) protected mice against infections by L. (L.) infantum and L. (L.) donovani. Sustained levels of IL-2 (against NH36 and SMT) and TNF-α (against NH36) were found in vaccinated mice, which exhibited reduced parasite load (COLER et al., 2015).

[082] Additionally, Leish-F3 induced increases in the frequencies of CD4+ T cells producing IFN-y, or IL-2 or TNF-α in PBMCs from patients infected with L. (L.) donovani from Bangladesh, and antibodies NH-specific IgG, higher in subclinical subjects than in human patients before treatment (COLER et al., 2015). Finally, Leish-F3 was also prepared under GMP conditions, being so far the only human subunit vaccine against human VL that has reached PHASE I testing in uninfected human adults in the USA, despite not having been licensed yet. . PBMCs from vaccinated subjects showed increased secretion of IFN-γ, TNF-α and IL-2 and low levels of IL-5 and IL-10 in response to Leish-F3. The vaccine was well tolerated (COLER et al., 2015). A patent application has been granted with the USPTO regarding this formulation (US20160186158A1). epitope vaccines

[083] The production and storage of recombinant proteins are often problematic and require good facilities, which limit their use in many contexts. On the other hand, peptides are more easily produced and shipped, and they can be stored lyophilized (LESTINOVA et al., 2017).

[084] The new vaccination approaches are based on the rational design of epitopes for B and T lymphocytes that promise to induce repertoires of immunological specificities, as well as to deal more safely with the genetic variation of both pathogens and humans. In addition, synthetic peptides are produced by chemical synthesis and are advantageous from a manufacturing point of view (OYARZUN & KOBE, 2016).

[085] More recently, the concept was developed that the most effective immune response against pathogens is derived from several different T cells that respond to a set of short peptides derived from pathogens called epitopes (DE GROOT et al., 2002). The theory of vaccines based on epitopes postulates that the potency and efficacy of a vaccine can be increased if, in order to optimize it, fragments that are smaller and smaller, but more immunogenic, of the selected protein are used.

[086] In this way, the mapping of epitopes of a protein can be used for the design of multiepitope, polyepitope or polytope vaccines (SAYED et al., 2016). Sets of polyepitopes are preferred to replace the living pathogen organism, whose use would always be linked to the risk of generating the disease. Indeed, the peptide epitopes of a protein, or of different proteins of a strain, or of conserved proteins of different strains of a species or genus, can now be easily assembled into a single polyvalent or universal vaccine (DE GROOT et al. , 2002; SAYED et al., 2016).

[087] The epitopes are selected according to their ability to bind in the clefts of histocompatibility receptor molecules (MHC in mice and HLA in humans) of antigen presenting cells (APC). These molecules present the epitopes to lymphocytes initiating the immune response. [088] The epitope anchor residues attach to pockets of the MHC molecule cleft, and the sequence is on the floor of this binding cleft, available to be recognized by lymphocyte receptors (CHRISTOPHER et al., 2013), which will trigger the immune response.

[089] The binding of the epitope to MHC-II can be considered a limiting factor for the subsequent exposure of the epitope on the cell surface. In addition to the accommodation of the side chain of amino acid residues in the “pockets” of the MHC-II binding cleft, studies demonstrate that hydrogen bonds are responsible for the discrimination between the charged and empty states of MHC-II, suggesting that the loss of a The single hydrogen bond between the main chain of the epitope and conserved MHC-II residues is capable of favoring the proteolysis of this complex (ARNESON et al., 2001, PAINTER et al., 2008; YANEVA et al., 2009).

[090] The antigenic peptide binds to MHC-II in a distended manner, being governed by hydrogen bonds resulting from the interaction between conserved residues present in the helices of the MHC-II cleft and the “skeleton” of the peptide. This conformation allows the accommodation of the side chain of amino acid residues in pockets P1, P4, P6/P7 and P9 which are called anchor residues, responsible for stabilizing the complex.

[091] The inherent properties of these pockets determine their specificity for certain residues defining the anchor amino acids of each MHC-II molecule isotype (JONES et al., 2006; PAINTER et al., 2008). As a result, their knowledge implies the rational development of peptides or proteins to be used as vaccine targets, for example (JAMES et al., 2009). There are, however, many alleles for each gene that encodes the proteins of the MHC and HLA receptors, and this variability generates structural differences in these proteins, which influence the binding with the epitopes (FLERI et al., 2017). This factor could hinder the development of a vaccine that should be protective for a genetically heterogeneous population with regard to its HLA molecules. [092] However, currently, in silico epitope prediction programs identify minimal sequences capable of binding to many MHC or HLA alleles. These epitopes, called promiscuous (FLERI et al., 2017), are the best candidates to compose a multi-epitope vaccine.

NH36, its domains and epitopes in the immune response of patients with visceral leishmaniasis

[093] Continuing with the search for the most immunogenic domains and sequences of NH36 for humans, we searched for the NH36 epitopes that would be presented by human HLA Class II and Class I molecules, using the epitope predictor bioinformatics programs TEPITOPE and SYFPEITHI . The aim was to identify epitopes that could bind to a greater number of histocompatibility molecules. A vaccine composed of these epitopes, called promiscuous, could generate broader protection in human populations without being restricted to protecting individuals who have some specific HLA alleles.

[094] In order to start the development of a human vaccine against leishmaniasis, this study identified 3 predicted epitopes in NH36, capable of binding to at least 19 of the 25 most frequent human HLA DR molecules: 1 epitope in the F1 of NH36, and two in F2, in addition to 8 epitopes for CD8 lymphocytes (3 in F1, 2 in F2, 3 in F3) (BARBOSA SANTOS et al., 2017). According to the prediction, for a human vaccine, therefore, the F1 and F2 domains would be the most important.

[095] Comparing the reactivity of individuals with VL, who are immunosuppressed, with that of individuals with asymptomatic infection or cured individuals, who have an active cellular immune response against Leishmania and are DTH+, would guide us in identifying the epitopes and domains to include in a human vaccine (KUMAR et al., 2012; SANTOS et al, 2017; STOBER et al., 2012). [096] Coincidentally with the predictions in silica, studying Brazilian patients infected with L. (L.) infantum chagasi, it was detected as markers of natural resistance to VL, the F2 domain, which induced the highest levels of IFN-y, IL -11β and TNF-α and, together with F1, the strongest secretion of IL-17, IL-6 and IL-10 by in vitro PBMCs from DTH+ and cured individuals (BARBOSA SANTOS et al., 2017). F2 also promoted the highest frequencies of IL-2 or TNF-a-producing CD4 ⁺ T cells, IL-2 and IFN-y, and IL-2, TNF-α and IFN-y producing multifunctional cells in cured individuals and asymptomatic DTH+.

[097] The increase in IFN-γ correlated with decreases in the size of spleens and livers and with increases in hematocrits in response to F1, and with increases in hematocrits and hemoglobin counts in response to F2. Additionally, increased IL-17 secretion in response to F1 and F2 was associated with decreased spleen and liver sizes.

[098] In contrast, the F1 and F3 domains increased the frequencies of CD8 ⁺ cells secreting IL-2, IL-2 and TNF-α, and multifunctional CD8 ⁺ cells secreting IL-2, TNF-α and IFN -y in patients with active VL. These increases were correlated with increases in the sizes of spleens and livers and decreases in hemoglobin and hematocrit counts (BARBOSA SANTOS et al., 2017).

[099] It was therefore concluded that healing and resistance to VL are related to a CD4+-Th1 and Th-17 response to the F2 and F1 domains and active VL is associated with CD8+ T responses against the F3 and F1 domains, likely involved in the initial infection control (BARBOSA SANTOS et al., 2017). Finally, and confirming the finding of predicted epitopes that bind to at least 19 of the 25 most common haplotypes of HLA-DR, HLA-A and HLA-B molecules, it was described that PBMCs from DTH+ individuals secreted IFN-y in response to the epitope synthetic predictors of the F1 domain, called SEQ ID NO:1, and in greater proportion, in response to the epitopes of the F2 domain, called SEQ ID NO:2 followed by SEQ ID NO:3 (BARBOSA SANTOS et al., 2017). [100] It was also described that the epitopes SEQ ID NO:31 , SEQ ID NO:2 and SEQ ID NO:3 of NH36, identified in 2014 using the TEPITOPE and SYFPEITH programs, are highly conserved in the genus Leishmania (BARBOSA SANTOS et al ., 2017). Additionally, in individuals infected with L. (L.) infantum from Spain, it was observed that the greatest lymphoproliferative response of patients cured of VL is directed against the F1 peptide (CARRILLO et al., 2017) and is accompanied by increased ratios of IFN-y/IL10 and TNF-α/IL-10 and IL-17 secretion, in asymptomatic individuals, and Granzyme B, in cured patients. It was concluded that the F1 domain is an inducer of the Th1 and Th17 response in cured and asymptomatic individuals infected with L. (L.) infantum (CARRILLO et al., 2017).

[101] Confirming the theory of epitope vaccines, it was possible to demonstrate that the use of increasingly smaller sequences of NH36, although immunogenic, potentiates and optimizes the immune response and generates a greater protective response, both in mice and dogs, as well as in humans . Additionally, also confirming that 3 of the epitopes predicted by the TEPITOPE program, and which bind to at least 19 of the 25 most frequent alleles of MHC class II HLA-DR molecules, stimulate the secretion of IFN-y by PBMC of DTH+ asymptomatic individuals.

[102] The present invention deals with the selection of genetically modified epitopes for the production of multiepitope proteins and the formulation of an immunogenic composition against leishmaniasis, designed with even more demanding criteria since the component epitopes were predicted with more sensitive algorithms, which link to more alleles of more histocompatibility genes. In fact, the prediction considered 50 of the most frequent alleles of MHC Class II molecules HLA-DR, DQ and DP. These promiscuous epitopes can be recognized by the widest range of individuals.

[103] The invention combines the promiscuous epitopes and uses an AAA sequence (Alanine-Alanine-Alanine) or a GPGPG sequence (Glycine-Proline-Glycine-Proline-Glycine) as spacers to prevent the epitope from being digested by the presenting cell of antigen, and avoid juxtaposed epitopes (LIVINGSTON et al., 2002), thus forming multiepitope proteins. Multifunctional T Lymphocytes

[104] The CD4+ effector and memory T cell differentiation model by (SEDER et al., 2008) starts with naive CD4 T lymphocytes, which after contact with a vaccine develop early secretion of TNF-α followed by IL- 2, of TNF-α and IL-2, and late secretion of IL-2, TNF-α and IFN-y simultaneously, by CD4+ cells called multifunctional or polyfunctional, which can persist as memory or effector T cells (CARRILLO et al., 2017).

[105] CD4 ⁺ T cells normally can remain as memory T cells or differentiate into effector cells, or they can die by apoptosis soon after their activation (SEDER et al., 2008). The existence of these distinct subsets of memory/effector and naive CD4+ T cells may play an important role in different clinical settings. Therefore, they need to be considered in immunological monitoring in vaccination and immunotherapy strategies (CACCAMO et al., 2018).

[106] The classification of memory T cell subsets is based on the expression of receptors necessary for extravasation through high endothelial venules to secondary lymphoid organs, such as CCR7+ / CD62L+ (SALLUSTO, 2016) and can be divided into at least three different populations: virgins/(naive) referred to as CD45RA+CCR7+, core memory as CD45RA-CCR7+, and effector memory as CD45RA-CCR7- (ORLANDO et al, 2016).

[107] The CC7 Chemokine Receptor (CCR7) is a membrane receptor, present on the surface of lymphocytes, that is involved in organizing thymic architecture and function, and in directing naive and regulatory T cells to various secondary lymphoid organs such as lymph nodes and spleen, it has also been shown that CCR7 has the ability to stimulate the maturation of dendritic cells and is involved in targeting T cells (SEDER et al., 2008).

[108] The study by Sallusto and Lanzavecchia (1999) showed that CD4+CCR7+ (central memory) cells express higher levels of IL-2, whereas CD4+CCR7- (effector memory) cells produce more IFN-y. [109] In addition, multifunctional CD4 ⁺ and CD8 ⁺ T cells whose frequency is increased in response to antigens have an essential role in protecting against a variety of infectious diseases caused by intracellular pathogens as has already been demonstrated in leprosy (SANTOS et al., 2018), and in tegumentary (NICO et al., 2014) and visceral (NICO et al., 2018) leishmaniasis.

[1 10] In this invention, the efficacy of the component epitopes of the multiepitope proteins and the immunogenic composition was evident when studying their ability to induce the secretion of cytokines, intracellularly and in the supernatants, by T CD4+ and T CD8+ lymphocytes from asymptomatic patients and DTH+ and patients cured of visceral leishmaniasis.

Detailed Description of the Invention

[1 1 1] To characterize the epitopes for potentially promiscuous CD4 ⁺ T lymphocytes, which bind most class II histocompatibility molecules, the complete sequence of NH36 (Pubmed Databank Protein AAG02281 ) was initially subjected to analysis by the prediction program PROPPED (http://crdd.osdd.net/raqhava/propred/) using a 3% cut off. PROPRED identified the most promiscuous 20-24 amino acid sequences, which bind to 75-88% of the 51 most frequent human HLA DRB1 molecules (Table 1).

[1 12] A previous test, carried out with the TEPITOPE program that analyzes binding to 25 molecules of HLA DRB1 , had identified only the epitopes 1 -2014, 2-2014 and 3-2013 (BARBOSA-SANTOS et al., 2017) . The PROPRED program considers binding to the 25 molecules considered by TEPITOPE plus 26 other molecules from the DRB1 gene alone.

[1 13] The IEDB program was also used, initially with a 5% percentile rank to identify in the NH36 sequence the epitopes that bind to the DQA1 VDQB1 * and DPA /DPB1 * genes. The IEDB algorithm confirmed two epitopes identified by PROPRED and 5 new epitopes. Finally, the previous results were confirmed, performing an analysis of the epitopes identified for HLA class II with the IEDB recommended 2.22 method, using 10% and 20% percentile rank. [1 14] This analysis predicted epitope binding to the 27 most frequent alleles of human HLA class II molecules, not only from DRB1 , DQA1/DQB1 and DPA1/DPB1 , but also from DRB3, DRB4 and DRB5 (Table 1 ).

[1 15] The peptides were synthesized by Genscript and named: 1 -2014 (SEQ ID NO:1 ), 2-2014 (SEQ ID NO:2), 3-2014 (SEQ ID NO:3), 1 -2018 ( SEQ ID NO:4), 3-2018 (SEQ ID NO:5), 5-2018 (SEQ ID NO:6), 7-2018 (SEQ ID NO:7), 9-2018 (SEQ ID NO:8) , 11-2018 (SEQ ID NO:9), 13-2018 (SEQ ID NO:10), 15-2018 (SEQ ID NO:11) and 17-2018 (SEQ ID NO:12), as shown in Table 1 . The reactivity of these epitopes with serum antibodies, the intracellular secretion of cytokines by CD4 ⁺ and CD8 ⁺ T cells and the secretion of cytokines in PBMC supernatants from asymptomatic and symptomatic human patients cured of VL were evaluated.

[1 16] Additional prediction using the IEDB recommended 2.22 program for 27 alleles additionally identified the “border line” epitopes BL-1 (SEQ ID NO:14) and BL2 (SEQ ID NO:15) (Table 1) , which were synthesized and used together with others in the in vitro interaction assay with recombinant HLA Class II molecules. Finally, the sequence of the 15-2018 and 17-2018 epitopes, which had a high overlap, were combined to form the new epitope 15-17-2018 (SEQ ID NO:13) (Table 1) which was also included in the immunoinformatics analysis .

[1 17] Subsequently, within the epitopes for class II, the sequences that bind to 27 molecules of human HLA-A and HLA-B class I were predicted. Analysis of the worldwide population coverage of epitopes for HLA class II and HLA class I alleles was also performed. Finally, we also identified which of the epitopes for HLA class II bind to DRB1 and DQA1 alleles considered to be associated with susceptibility or resistance. the human VL. Thus, HLA typing was performed on the 10 cured patients and 10 DTH+ patients studied in Sergipe, at the HLA-UERJ Laboratory. [1 18] Finally, with all the information gathered, a multiepitope protein was designed using genetically modified epitopes (artificial sequences) through the addition of amino acid spacers, to be used in vaccination, treatment, transfusion control and diagnosis of human leishmaniasis and animals.

Table 1. NH36 epitope prediction for HLA Class II molecules. Sequences of the epitopes containing the amino acids predicted by the programs PROPRED for molecules DRB*1 , and IEDB for molecules DQA17DQB1 * and DPA17DPB1 *, DRB*1 , DRB*3, DRB*4, DRB*5.

Epitope testing in human VL patients

[1 19] For this invention, we identified the predicted epitopes of NH36 that can bind to MHC class II molecules DR, DB and DQ. These molecules present antigens to CD4+ T lymphocytes that are involved in protection against VL, keeping the cellular immune response to the parasite intact. VL determines a severe loss of cellular immune response to Leishmania. To guide the construction of a preventive multiepitope immunogenic composition (vaccine), sera and PBMC from groups of patients with natural resistance to VL and preserved CD4+ T immune response (asymptomatic DTH+) and patients cured of VL were used. these responses were compared with those of healthy controls from a non-endemic area.

[120] The antigenicity of the epitopes was tested for serum antibodies, through ELISA assay, and their abilities to induce production and secretion of cytokines, through the LUMINEX assay and flow cytometry. Samples of patients with visceral VL diagnosed and treated at the University Hospital of Sergipe by Dr. Roque Pacheco de Almeida were obtained. Peripheral blood samples provided by patients cured of Visceral Leishmaniasis (n=10) and from relatives of patients who were asymptomatic but were positive for the Montenegro test (DTH+) (n=10) and healthy controls (n=10) were used. ).

[121] In order to carry out this research, all patients signed an informed consent form (TCLE), confirming their participation. The use of these cells has already been submitted and approved by the ethics committee of the HU/UFS under CAAE-0123.0.107.000-1 1 . The criteria used to classify the clinical forms of VL are described below:

[122] Asymptomatic: family members and contacts of patients who underwent the delayed-type hypersensitivity test (DTH+) for soluble Leishmania antigen and presented a papule greater than or equal to 5mm on the surface of the forearm, in addition to which they did not present clinical signs or symptoms of VL classic.

[123] Cured: Individuals who have completed VL treatment and who have not relapsed within a 180-day period, aged 12 to 50 years. Alcoholic patients, children, co-infection with HIV, Hepatitis and other pathologies of infectious origin were excluded from this study.

[124] Additionally, for serological assays, control samples from a non-endemic area obtained from healthy blood donors from the Clementino Fraga Filho University Hospital of the Federal University of Rio de Janeiro were provided. The sera used showed negative responses in tests for HIV, Chagas Disease, Hepatitis B and C, Syphilis, HTLV and Chagas Disease.

[125] The "buffy coats" of seronegative blood donors used for ICS tests were also obtained from the Blood Bank of the Clementino Fraga Filho University Hospital of the Federal University of Rio de Janeiro, a project of the Oswaldo Cruz Institute of Rio de Janeiro (CAAE 35775014.0.0000.5248), approved on April 24, 2018, with the collaboration of FIOCRUZ, RJ. Predicted synthetic NH36 epitopes are recognized by antibodies of the type

IgG, lgG1, lqG2, lqG3 and lqG4

[126] Synthetic NH36 epitopes were tested in an ELISA assay against sera from healthy controls from a non-endemic area, asymptomatic DTH+ individuals, and cured VL patients (Figure 1). Significant differences were detected between the responses of healthy controls from a non-endemic area and cured (p < 0.0001 ) and DTH+ (p < 0.0001 ). In general, the responses of cured individuals were similar to those of DTH+ individuals. The highest absorbances of IgG antibodies were recorded in response to epitopes 1 -2014 (SEQ ID NO:1 ) and 2-2014 (SEQ ID NO:2) (Figure 1 ).

[127] The lgG1 response was also strong against sequence 1-2014 and epitopes 3-2018 (SEQ ID NO:5) and 5-2018 (SEQ ID NO:6). However, it was the epitope 15-2018 (SEQ ID NO:1 1 ) that predominated in the lgG2 response, followed by 7-2018 (SEQ ID NO:7), 9-2018 (SEQ ID NO:8) and 1 -2014. In the lgG3 response, the epitopes 3-2018 (SEQ ID NO:5) and 7-2018 (SEQ ID NO:7) stood out, followed by the other epitopes of 2018 and the sequence 2-2014. Finally, IgG4 antibodies were more directed against epitopes 3-2014 and 3-2018 followed by 7-2018 and 9-2018 (Figure 1).

[128] Additionally, the representation of the analysis results by ELISA through the “heatmap” facilitated the graphic distinction of the most potent epitopes and the types of antibodies in higher concentration that determined higher absorbances. Indeed, in Figure 2 it is possible to confirm that the highest absorbances were developed by the IgG antibodies, followed by a less intense response to the IgG 1 , IgG2 and IgG3. The lowest absorbances were detected for lgG4.

[129] The heatmap confirms that epitopes 1 -2014 and 2-2014 are the most potent in the IgG response, with 1 -2014 also eliciting a strong IgG response 1 . On the other hand, the 3-2018 and 5-2018 epitopes induce the lgG1 response and, together with the 7-2018, the lgG3. The predominant in the lgG2 response was the epitope 15-2018 with similar intensity to epitopes 1 -2014 and 2-2014 for IgG. The lgG4 response was most potent against the 3-2014 and 3-2018 epitopes (Figure 2). [130] The results reveal the antigenic potential of the NH36 epitopes predicted for the HLA-class II system, in the recognition of antibodies generated during infection by L.(L.) infantum chagasi. One of the symptoms of VL is hypergammaglobulinemia with high levels of circulating anti-Leishmania antibodies, which, however, are not protective. However, a vaccine that induces, in addition to a cellular response, a strong humoral response can bring the benefit of blocking transmission by the insect vector in endemic areas, as described for TBV vaccines against Malaria (WHO, 1997).

[131] The Leishmune® vaccine is, for example, a transmission blocking vaccine (TBV) and the antibodies generated by it contribute to the interruption of the infectious cycle in nature (PALATNIK DE SOUSA et al., 2009, SARAIVA et al., 2006 , PALATNIK DE SOUSA et al., 2008). Elevated levels of anti-Leishmania lgG1 are associated with therapeutic failure or relapse of the disease in post-dermal kala-azar, whereas after healing, lgG1 levels decrease (MARLAIS et al. 2018).

[132] Coincidentally, elevated levels of anti-Leishmania (L.) donovani lgG3 and lgG1 antibodies have been described in patients with post-dermal kala-azar (SAHA et al., 2005), and considered IL-10 surrogate markers in VL , because they are concomitant with high levels of IL-10 (GANGULY et al., 2008) and decrease with the healing time of cutaneous leishmaniasis (FAGUNDES SILVA et al., 2012).

[133] In addition, ANSARI et al., (2008) described that anti-Leishmania antibodies lgG3 and lgG4 were significantly increased in children with VL whereas IgG, lgG1 and lgG2 levels were comparable in pediatric and adult cases of VL post -dermal. ANAM et al., 1999, define lgG3 as the subclass related to VL marker and SAHA et al., (2005) described high levels of IgG, lgG1, lgG2, and lgG3 antibodies found in patients with post dermal kala-azar, but not with VL from India (SAHA et al., 2005). [134] GOSH et al., (1995) found in patients with VL from India, levels of anti-Leishmania antibodies lgG1 > lgG2 > lgG3, and very little lgG4. Additionally, antibodies of the lgG3 isotype reacted with antigens in the 14 to 34 kDa range and persisted for several months after cure (GOSH et al., 1995).

[135] In all these works, antibody subtypes were not detected completely correlated with the cure or resistance to VL, which emphasizes the strong immunogenicity of the NH36 epitopes, which are recognized by IgG, lgG1, lgG2, lgG3 antibodies and, in lower concentrations by lgG4 antibodies from cured and asymptomatic patients (Figure 2).

[136] In this sense, it is worth noting that NH36 was strongly recognized by IgG antibodies in sera from VL patients and asymptomatic individuals from Bangladesh, while there was a preferential increase in IgG, IgG 1 and IgG3 antibodies in individuals vaccinated with a chimera recombinant containing NH36 (COLER et al., 2015). The description of the different affinity of the NH36 epitopes for antibodies from cured patients and DTH+ helped in the design of an immunogenic composition (vaccine) with multiepitope proteins targeting the cellular and humoral response against VL.

Predicted synthetic NH36 epitopes induce intracellular expression of cytokines in CD4+ T lymphocytes

[137] To assess the production of pro-inflammatory intracellular cytokines in TCD4+ and TCD8+ lymphocytes from healthy controls, patients cured of VL and asymptomatic individuals, the technique of multiparametric flow cytometry and Boolean analysis was used, which consists of using different fluorochromes for the detection of cytokines expressed alone, or the expression of two or three cytokines simultaneously.

[138] Mainly, the so-called triple-secretory or multifunctional T cells are related to the immunological memory that one wants to achieve with a preventive vaccine. For this, the gate strategy was used, as can be seen in Figure 3. [139] Initially, the population of viable cells was separated in order to exclude excess “debris” as well as dead cells and within this population lymphocytes were selected using the parameters size by granularity (FSC x SSC). From this population of lymphocytes, only those that were CD3+ and subsequently CD3 ⁺ CD4 ⁺ and CD3 ⁺ CD8 ⁺ were selected.

[140] Considering the evaluation of only the population of CD3 ⁺ CD4 ⁺ and CD3 ⁺ CD8 ⁺ lymphocytes that had contact with the epitopes, and became effector and memory lymphocytes, to reflect a better response to the studied antigens, the population of naive lymphocytes was excluded (CCR7 ⁺ CD45 ⁺ ). Thus, the lymphocytes included in the analysis are effector (CD45+CCR7-) and/or memory (CD45-CCR7+) cells, which were those that had contact with the antigens and are able to fight the infection.

[141] Finally, the production of cytokines IL-2, TNF-α, IFN-Y was analyzed in them, and then Boolean analysis was performed to detect the simultaneous production of two cytokines (IL-2+TNF- α+, IL-2+IFN-Y+ and TNF-O+IFN-Y+) OR three cytokines, in triple-secretory or multifunctional lymphocytes (IL-2+TNF-α+IFN-Y+) by the effector population and memory.

[142] The percentage of TCD3+ lymphocytes stimulated with the epitopes of NH36 (1 -2014 (SEQ ID NO:1 ), 2-2014 (SEQ ID NO:2), 3-2014 (SEQ ID NO:3), 1 - 2018 (SEQ ID NO:4), 3-2018 (SEQ ID NO:5), 5-2018 (SEQ ID NO:6), 7-2018 (SEQ ID NO:7), 9-2018 (SEQ ID NO: 8), 11-2018 (SEQ ID NO:9), 13-2018 (SEQ ID NO:10), 15-2018 (SEQ ID NO:11) and 17-2018 (SEQ ID NO:12), as well as with the soluble Leishmania donovani antigen (SLA) is similar in the healthy control, cured VL (D180) and asymptomatic (DTH+) groups. (Figure 4A).

[143] Within this population of CD3+ T lymphocytes, the percentage of CD3 ⁺ TCD4 ⁺ lymphocytes averages 50% (Figure 4 B) and CD3 ⁺ CD8 ⁺ lymphocytes (Figure 4 C) approximately 40%, in the latter case with slight increase in cured individuals. [144] Since there was no difference in the percentage of lymphocytes between healthy controls, patients cured of VL and asymptomatic individuals, the evaluation of the immune response was carried out through the analysis of the production of intracellular cytokines, after stimulation with NH36 epitopes.

[145] Regarding the intracellular production of a single cytokine (Figure 5) it is worth noting that the frequencies of CD4+ T lymphocytes producing IL-2, TNF-α or IFN-y were predominantly higher in DTH+ individuals than in cured individuals , and higher in these than in controls, with the exception of the 9-2018 epitope, which increased the proportions of TCD4+TNF-α+ lymphocytes more in controls (Figure 5B). It is also important to highlight that all epitopes are more potent than the soluble SLA antigen in increasing the proportions of IL-2 or IFN-y producing CD4+ T lymphocytes (Figure 5 A and 5 C). Epitopes 3, 9 and 11 -2018 stood out in the increase in frequencies of TCD4+IL-2+ lymphocytes (Figure 5 A); epitopes 1 , 7 and 13 -2018 for the frequencies of TCD4+-TNF-α cells (Figure 5 B), and epitopes 5, 7, 9 and 17-2018 for the proportions of TCD4+IFN-Y cells (Figure 5 W).

[146] Additionally, healthy controls had higher frequencies of dual-producing IL-2 and TNF-α CD4+ T cells than cured or DTH subjects (Figure 6A). This was most notable in the case of the 9-2018 epitope which had already been shown to induce increases in controls in the frequencies of TNF-α producing CD4+ T cells alone. In contrast, TNF-a and IFN-y producing TCD4+ lymphocytes had their frequencies increased in DTH+ individuals, mainly in response to epitopes 1 -2014 and 5, 7, 9 and 11 -2018 (Figure 6 B).

[147] SLA antigen alone raised these proportions in cured patients. Finally, the frequencies of TCD4+ lymphocytes producing IL-2 and IFN-y were also increased in DTH+ individuals, mainly in response to epitopes 2-014 and 3 to 17-2018 (Figure 6 C). [148] Finally, it was possible to detect increases in the frequency of triple secretory or multifunctional TCD4+ lymphocytes, mainly in DTH+ individuals, promoted by most epitopes, with emphasis on epitopes 9 and 11 - 2018 followed by sequences 1 -2014, and 3, 5, 7 and 15 of 2018. The response to the SLA antigen is greater in cured individuals (Figure 7), as well as detected in lymphocytes secreting TNF-α+ and IFN-y+ (Figure 6 B).

[149] Additionally, a heatmap was generated based on the average frequency values of cytokine-secreting CD4+ T lymphocytes (Figure 8) to help identify the most potent epitopes. In this case, we used a graph normalized by columns to facilitate the visualization of the most potent epitopes, even for cell subtypes that appear less frequently, such as triple-secretors and multifunctional ones. The conclusions are summarized in Table 2.

[150] It was then concluded that, although several epitopes showed great potential, it was epitopes 9, 11 , 15 and 3-2018 that were able to increase the frequency of multifunctional cells probably related to immunological memory to the NH36 antigen (Figure 8 and Table 2).

Table 2. Most potent epitopes in increasing the frequencies of cytokine-secreting CD4 T lymphocytes.

Predicted synthetic epitopes of NH36 for CD4+ T lymphocytes also induce intracellular expression of cytokines in CD8+ T lymphocytes

[151] Additionally, the CD3+CD8+ T lymphocyte response was studied after incubation with the predicted epitopes for synthetic CD4+ T lymphocytes from NH36. With respect to the intracellular production of a single cytokine, it was observed that the frequencies of IL-2 or IFN-y producing CD8+ T lymphocytes were, as well as detected for TCD4+ cells, predominantly higher in DTH+ individuals than in cured and controls. The proportions of IL-2 producing CD8+ T cells were increased in response to epitopes 9, 11 and 17-2018, followed by 2-2014, and 3 and 15-2018 (Figure 9A). In the case of TNF-α production, epitopes 9 and 1 1 stood out, increasing by 84% and 60%, respectively, the proportions of TCD8+TNF-α+ cells in DTH+ individuals (Figure 9 B). However, controls also responded with low frequencies against most epitopes. Finally, the frequencies of TCD8+IFN-Y+ lymphocytes were more pronounced in response to epitopes 7, 9 and 17 -2018, followed by epitopes 1 -2014, 3, 5 and 1 1 -2018 (Figure 9C).

[152] Afterwards, the simultaneous production of two cytokines in TCD8+ lymphocytes was studied (Figure 10). Joint secretion of IL-2 and TNF-α was higher in cured and DTH+ patients, with the exception of epitopes 7, 9 and 13-2018 in which the frequencies of controls were increased (Figure 10 A). The most potent epitopes were 3-2014 and 9-2018. In contrast, TNF-α and IFN-y producing CD8+ T lymphocytes (Figure 10 B), and IL-2 and IFN-y (Figure 10 C) had their frequencies increased in DTH+ individuals, mainly in response to epitopes 9 and 1 1 - 2018, in TNF-α and IFN-y producing cells (Figure 10 B), and to epitopes 5, 9, 15 and 17-2018, in IL-2 and IFN-y producing cells (Figure 10 W).

[153] Finally, and as described for CD4+ T lymphocytes (Figure 7), it was possible to detect increases in the frequencies of triple-secretory or multifunctional TCD8+ lymphocytes, mainly in DTH+ individuals, promoted by most of the epitopes, with emphasis on the epitopes 9 and 11 -2018 followed by sequences 1 - 2014, and 3, 7 and 15 of 2018 (Figure 11).

[154] The response to the SLA antigen was, as described for multifunctional CD4+ T lymphocytes (Figure 7), greater in cured individuals, both in multifunctional CD8+ cells (Figure 11) and in IL-2-secreting CD8+ T lymphocytes (Figure 9A ), IL-2 and TNF-α+ (Figure 10 A) and TNF-α+ and IFN-γ+ (Figure 10 B). [155] Confirming the immunogenic potency of the epitopes, it is worth noting that the SLA control always worked by inducing smaller or similar responses to the epitopes, except in CD4+ and CD8+ T cells that double produce TNF-α and IFN-y, in which it induced a response greater than the epitopes, in cured individuals.

[156] A heatmap (Figure 12) was also generated based on the average frequency values of TCD8+ cytokine-secreting lymphocytes (Figure 9-11) to aid in the identification of epitopes. We also used a graph normalized by columns for CD8+ T lymphocytes, as we did for TCD4+ lymphocytes (Figure 8) (Figure 12). The conclusions are summarized in Table 3.

[157] It was then concluded that, although several epitopes showed great potential, it was epitopes 9, 11 , 15 that were able to increase the frequency of multifunctional TCD8+ cells probably related to the immunological memory to the NH36 antigen.

Table 3. Most potent epitopes for increasing the frequencies of secretory CD8 T lymphocytes

Predicted synthetic NH36 epitopes induce cytokine secretion by PBMC

[158] It was observed that the DTH+ group had the best response for the production of multifunctional T lymphocytes. Then, the secretion of pro-inflammatory, anti-inflammatory and regulatory cytokines in PBMC supernatants stimulated with NH36 epitopes from the group of DTH+ individuals was evaluated.

[159] For this purpose, the PBMC culture supernatant was collected at 72 hours and the levels of cytokines IL-2, TNF-α, IFN-y, IL-12p70, IL-1 β, IL-4, IL-6 , TGF-β, IL-22, IL-10 were measured using the Luminex system. It should be noted that as a control Baseline cells from each DTH+ individual were used, without stimulation, as a comparison parameter. The results are summarized in Figure 13.

[160] With regard to the induction of pro-inflammatory response, it was possible to observe that the NH36 epitopes induce a significant increase in the production of the cytokine IL-113 in response to the 1 1 -2018 epitope, as well as to the 13- epitopes 2018, 9-2018, 2-2104 and 1 -2018, with levels greater than those observed in the baseline control (33 pg/ml) and in the SLA antigen (108 pg/ml) (Figure 13). An increase in TNF-α secretion in response to the 9-2018 epitope (265 pg/ml) was also observed when compared to the basal control (136 pg/ml) (Figure 13). As for the production of IL-12p70, there was a slight increase in the production of this cytokine in response to the 3-2014 epitope (1 .3 pg/ml) when compared with the other epitopes and with the basal control (0.5pg/ml) (Figure 13 ). Finally, the production of cytokines IFN-y and IL-2 was observed only in stimulation with the SLA antigen. Only the 3-2014 epitope promoted increases in IFN-y and IL-2 secretion, but at levels similar to baseline control (Figure 13).

[161] Secretion of the anti-inflammatory cytokine IL-6 was highest in response to the 11 -2018 epitope (1600 pg/ml), whereas controls secreted approximately 400 pg/ml (Figure 13). Additionally, there was also an increase in TGF-[3 production for epitopes 1 -2014, 2-2014 and 13-2018 (49 to 61 pg/ml), twice the values found for SLA and for the basal control (approximately 20 pg/ml) (Figure 13).

[162] On the other hand, IL-10 production was observed only with SLA stimulation and in the basal control group, whereas IL-22 was only secreted by the SLA antigen. There was no production of the cytokine IL-4 in response to the NH36 epitopes (Figure 13).

[163] In conclusion, with regard to the pro-inflammatory response, epitope 9 in TNF-α secretion stands out, and epitope 11 along with 9, 7, 1, 3, 13 and 15 of 2018 and 2, 3 de 2014 on IL1 β secretion. In addition to IL1 β, the 2018 epitope 11 also promotes IL-6 production. Additionally, epitope 13 from 2018, together with 1 and 2 from 2014 promote the production of TGF-[3. [164] Nucleoside hydrolase (NH36) is an important phylogenetic marker of the genus Leishmania that presents highly conserved domains among different species (NICO et al., 2017). In vivo studies in a murine model using immunization with the recombinant NH36 protein and its peptide fractions suggest an association with a protective immune response in VL (AGUILAR-BE et al., 2005; GAMBOA-LEÒN et al., 2006; NICO et al. ., 2017; NICO et al., 2018) and a cross-preventive immune response against Leishmania (L.) amazonensis (NICO et al., 2014, ALVES-SILVA et al., 2019) and Leishmania (V. ) braziliensis (ALVES-SILVA et al., 2019).

[165] In a previous study by this research group, SANTOS et al. (2017) described in Brazilian patients infected with L. (L.) infantum chagasi, as markers of natural resistance to VL, the F2 domain induced the highest levels of IFN-y, IL-1 p and TNF-α and, together with the F1, the strongest secretion of IL-17, IL-6 and IL-10 in DTH+ and cured individuals (SANTOS et al., 2017). F2 also promoted the highest frequencies of T cells CD4 ⁺ IL-2 ⁺ , CD4 ⁺ IL-2 ⁺ TNF-α ⁺ , CD4 ⁺ IL-2 ⁺ IFN-y ⁺ and CD4 ⁺ IL-2 ⁺ TNF-α ⁺ IFN -γ ⁺ in cured and asymptomatic individuals.

[166] In contrast, the F1 and F3 domains increased the frequencies of CD8 ⁺ IL-2 ⁺ , CD8 ⁺ IL-2 ⁺ TNF-α ⁺ and CD8 ⁺ IL-2 ⁺ TNF-α ⁺ IFN-y ⁺ cells in patients with active LV. In this study, Santos et al. concluded that VL cure and resistance are related to a CD4+-TM and Th-17 response to the F2 and F1 domains, and active VL is associated with CD8+ T responses against the F3 and F1 domains, probably involved in controlling the initial infection.

[167] Using the epitopes predicted by the TEPITOPE program, which binds to 25 molecules on HLA-DR, and SYFPEITH HLA-A and HLA-B, PBMC, DTH+ individuals secreted more IFN-γ in response to the synthetic epitopes 2-2014 and 3-2014 of the F2 domain followed by the epitope 1 -2014 located in the F1 domain [26].

[168] In the present invention, we used the epitopes 1 , 2 and 3-2014, but we also included in the study, the synthetic epitopes 1 , 3, 5, 7, 9, 11 , 13, 15, 17-2018, predicted by the Programs Propred and IEDB algorithms that include 50 haplotypes of MHC class II DR, DP and DQ molecules. [169] In this invention, therefore, a greater number of epitopes were used that represent the entire sequence of NH36 and that could be considered more promiscuous.

[170] NH36 epitopes induced a Th1-type protective phenotype mediated by production of the isolated cytokines IL-2, which is related to cell proliferation, as well as TNF-α and IFN-y, which are involved in controlling Leishmania infection , in addition to being related to the ability to generate immunological memory through the development of TCD4+ and TCD8+ lymphocytes of effector memory and central memory.

[171] Similar results were observed by CARRILLO et al., (2017) where the F1 domain of the NH36 protein suggested the development of a Th1-type protective response through the secretion of pro-inflammatory cytokines in PBMCs from patients cured of VL and in asymptomatic individuals. Likewise, Santos et al, described a robust response of IL-2 and TNF-α generated by NH36 domains.

[172] According to SEDER (2008), although IL-2 has little effector function, it promotes T cell expansion in an autocrine or paracrine manner, which could enhance the memory function of CD8+ T cells, as well as to amplify multifunctional T cell responses.

[173] In the present study, both PBMCs from asymptomatic individuals and cured patients (D180) produced a high percentage of pro-inflammatory cytokines when stimulated with NH36 epitopes, which indicates that previous contact with the parasite potentiates the development of a protective response.

[174] It was shown in this study that the NH36 epitopes: 3-2018, 9-2018, 1 1 -2018 and 15-2018 were the most immunogenic in the multiparametric analysis of ICS, as they induced a potent CD3 multifunctional TCD4+ lymphocyte response ⁺ CD4 ⁺ IL-2 ⁺ TNF-α ⁺ IFN-Y, mainly in asymptomatic individuals, followed by the group of cured VL patients. [175] Additionally, the 9-2018, 11-2018, and 15-2018 epitopes also enhanced the response of CD3 ⁺ CD8 ⁺ IL-2 ⁺ TNF-a ⁺ IFN-y multifunctional TCD8+ lymphocytes in DTH+ and cured patients.

[176] It is noteworthy that the percentage of these lymphocytes was twice as high as the values found by BARBOSA SANTOS et al., (2017) when using the NH36 domains as antigens. With this, it was proven that the identification and use of the most immunogenic epitopes potentiates the response above that generated by the total protein or its domains (PALATNIK de SOUSA, SOARES & ROSA, 2018, DE GROOT et al., 2002, SEYED et al. , 2016).

[177] This can be explained by the fact that the epitopes have shorter amino acid sequences that are, however, more restricted to the HLA class II receptor, thus generating a more potent response. This effect has already been confirmed with the NH36 vaccine in mice (NICO et al, 2010, 2014; Alves-Silva et al. 2017, 2019; PALATNIK de SOUSA, 2019). Thus, epitopes 3, 9, 11 and 15-2018 should be included in a synthetic vaccine aimed at inducing a Th1 and TCD8 cellular immune response. According to OYARZUN & KOBE (2016), the versatility to genetically arrange large combinations of capable B and T cell epitopes allows to induce robust humoral and cellular immune responses.

[178] Elevated TNF-α and/or TNF-α and IL-2 production was initially observed in some healthy controls. This was not expected in healthy controls from a non-endemic area with a negative serological test for Chagas disease. These high frequencies of TCD4+ and TCD8 lymphocytes that produce TNF-α and TNF-α and IL-2 were mainly detected in response to the 9-2018 epitope, which was one of the most potent in inducing multifunctional cells in cured patients and DTH+.

[179] It is possible that epitope 9 has an extremely strong ability to induce TNF-α alone and/or TNF-α and IL-2 and thus we observe this manifestation in controls. This result can be explained by the high quality of the antigens, capable of inducing TNF-α in this group. [180] It is also possible that the MHC class II DR, DQ, or DP and HLA alleles of some of the healthy control individuals have a higher affinity for epitope 9 determining this response in individuals who have not had contact with the parasite. The sample of individuals used in this study is relatively small and they have not yet been HLA typed.

Affinity of NH36 epitopes for HLA DR, DQ and DP molecules in silico and in vitro

[181] Prediction of the affinity of NH36 epitopes for the 27 most frequent human molecules from the DRB*1 , DRB*3, DRB*4, DRB*5, DQA*1/DQB*1 and DPA*1/DPB*1 genes is summarized in Figure 14. In a more restricted analysis (< 10% percentile rank), the epitopes 15-17-2018, 1 1 -2018 and 17-2018 are the ones that bind to a greater number of molecules and are, therefore, considered the most promiscuous. The broader analysis (< 20% percentile rank) confirms these epitopes and also incorporates sequences 2-2014, 5-2018, 9-2018 and 15-2018 (Figure 14).

[182] In addition to immunoinformatics prediction (Figure 14), affinity of NH36 epitopes for MHC receptors has been studied in vitro, based on their ability to inhibit binding of a radioactive peptide probe to purified MHC molecules (Sidney et al. , 2013. Curr Protoc Immunol. 2013 Chapter 18:U n it 18.3. doi: 10.1002/0471142735.im1803s100). We included in this analysis the epitopes 1, 2 and 3 from 2014; 1, 3, 5, 7, 9, 1 1 , 13, 15 and 17 of 2018 and two additional “Border Line” epitopes (BL1 and BL2), which were not included in previous immunological assays.

[183] Figure 15 depicts the IC50 ηM values developed by the epitopes when binding to 27 different HLA Class II molecules DPA1/DPB1, DQA1/DQB1, and DRB1, DRB3, DRB4, and DRB5. IC50 concentrations < to 1000 M, represented in bold, indicate the highest affinities of an epitope for each HLA receptor. Negative signs indicate IC50 values >30000 ηM (Figure 15).

[184] It is important to note that most epitopes are promiscuous and bind to at least seven (25%) of the 27 HLA class II molecules tested (Figure 15). Only the DRB1*12:01 allele did not bind to any epitope. [185] In contrast, 5 epitopes (11 -2018, 1 -2014, 5-2018, 13-2018 and BL1 ) bind to 13 (48% of alleles) and 1 epitope (2-2014) binds to 17 HLA molecules (63%). According to this approach, the epitopes 17-2018 and 15-2018, which bind to only 2 and 5 alleles, respectively, could be excluded from the vaccine, as they are not promiscuous.

[186] Table 4 shows the sum of the frequencies of all cytokine-secreting CD4+ and CD8+ T cells and the frequencies of triple-secreting multifunctional cells in response to the epitopes. The highest frequencies of all cell types were observed in response to the epitope 1 1 -2018, followed by 9-2018. The 1 1 -2018 epitope would therefore be not only more promiscuous (Figure 14 and 15) but also more immunogenic (Table 4).

Table 4. Number of HLA class II alleles that bind NH36 epitopes, sum of frequencies of all cytokine-secreting CD4 and CD8 T cells and IL-2TNF-olFN-y secreting cells in response to epitopes

[187] In Table 5 it is possible to compare the number of alleles predicted for the binding of each epitope, with 10% and 20% percentile rank, and the number of alleles actually bound in vitro. There are similarities between prediction and in vitro assay results for epitopes 1 1 -2018, 2-2014, 5-2018, 13-2018 and BL1 , which are the most promiscuous. On the other hand, the epitopes 15-2018 and 17-2018, despite being predicted to bind 1 1 and 21 alleles, effectively bound only 5 and 2 alleles, respectively. This result may be based on the high hydrophobicity of these epitopes, which could interfere with their binding capacity in vitro. Table 5. Prediction of allele linkage by immunoinformatics and in vitro linked alleles

[188] On the other hand, the 17-2018 epitope could be considered non-essential since it binds to the DQA1*05:01 / DQB1*02:01 allele, to which another 10 epitopes (9-2018, 1 1 -2018, 15 -2018, 2-2014, 1 -2018, 5- 2018, 13-2018, 7-2018, BL1 and BL2), and to the allele DRB3 * 01 : 01 , to which also, another 4 epitopes bind (11-2018, 2-2014, 13-2018, BL1) (Figure 15). Therefore, nothing would be lost by deleting the 17-2018 epitope. However, analysis of the immune response of previously studied human PBMC revealed that both the 15-2018 and 17-2018 epitopes increased the total frequencies of cytokine-secreting CD8 T cells, and the 15-2018 epitope was one of the four most potent in increasing CD4 + and CD8 + multifunctional T cell frequencies (Table 4).

[189] It is important to consider, however, that the sequences of the 15-2018 and 17-2018 epitopes overlap by 16 amino acids. In this way, it is possible to design a new epitope by extending the sequence of the 15-2018 epitope to also include the last four amino acids of the 17-2018 epitope (RIGD). This will not increase coverage, but it can increase coverage redundancy, which can be important if any leaks occur. Thus, in the immunogenic composition (vaccine) with multiepitope proteins, the epitopes 15-2018 and 17-2018 will be replaced by a new epitope called 15-17-2018, which is composed of the sequence QVAVKLDFDKFWCLVIDALKRIGD (Table 1). Selection of epitopes according to their coverage of class HLA molecules

[190] In Figure 15 it is possible to observe that the epitope 13-2018 is the only one that binds to the DRB1 *03:01 allele and, along with BL1 , they are the only two that bind to the DRB3*02:02 allele. Thus, both the 13-2018 and BL1 epitopes, which bind to 13 alleles, are also essential in the immunogenic composition (vaccine). In fact, 6 alleles bind to 3 epitopes simultaneously, 7 alleles to 4 epitopes, 2 alleles to 5 epitopes, 1 allele to 7 epitopes, 2 alleles to 9 epitopes, 1 allele to 10 epitopes, 1 to 1 1 epitopes and 1 allele to 13 epitopes.

[191] With regard to their antigenic activities, the epitopes 9-2018, 1 1 - 2018, 7-2018, 5-2018, and 3-2018, in descending order, induced the highest sums of CD4 ⁺ and CD8 T cell frequencies ⁺ secretory cytokines (Table 4). Accordingly, the epitopes 9-2018 and 1 1 -2018, followed by 15-2018 and 3-2018, and to a lesser extent by 1 -2014, 5-2018 and 7-2018 (Table 4), in descending order, were the most potent promoters of CD4 ⁺ and CD8 ⁺ multifunctional T cell frequencies (Table 4).

[192] Another criterion for estimating the best candidate epitopes is their potential population coverage, ie, their ability to bind to a wide range of alleles of HLA class II molecules. In Figure 15, the columns with two alleles show the alpha and beta chains for DQ and DP molecules, because for DQ and DP the alpha chain is polymorphic. In contrast, DRA is largely monomorphic (about 98% DRA1 *01 :01 ) and this variant does not influence binding. Hence it is common to ignore the DRA annotation and just annotate the beta strand variant with respect to DR (DRB).

[193] For DQ, the alpha chain definitely can affect binding specificity, so its annotation is important. That is, DQA1 *0301 /DQB1 *0301 has a different specificity than DQA1 *0501 /DQB1 *0301 . At the same time, the link between DQA and DQB is quite close, so you can generally infer what the alpha chain will be if you know the DQB, with a few exceptions (such as DQB1 * 0301 ). [194] For DP receptors, on the other hand, the alpha chain helps to form the binding groove, but it is not very polymorphic and does not have much effect on binding specificity. Indeed, from a peptide binding point of view, it is not possible to tell the difference between, for example, DPA1 *0103/DPB1 *0402 and DPA1 *0301/DPB1 *0402. However, as polymorphism can affect TCR recognition, DPA is usually annotated as well. For example, the correlation between DPA1 *0103/DPB1 *0402 and DPA1 *0301 /DPB1 *0402 for a hypothetical protein is 0.99, both for IC50 nM and rank, suggesting that there is no impact on DPA variation.

[195] With respect to DP molecules, DPB1 *04:01 and DPB1 *04:02 are the two most prevalent alleles in all populations of the world and in all major ethnic and geographic groups (AFND; www.allelefrequencies.net ;Requena et al.2020, Front.Immunol.11,2008,doi:10.3389/fimmu.2020.02008). For some populations, 0401 is most prevalent, while for others, 0402 is most frequent. One or the other is present in more than 60% of the human population. Both run about 35-40% on average in the general population, nearly double the next most common DP. In comparison, the most common DR will have an average frequency of around 20%. However, although the population coverage of these DP alleles is predicted to be very high, it is important to note that the cell surface expression of DP is considered to be somewhat lower than that of DQ and DR. Perhaps for this reason, DP is generally associated with less defined HLA-restricted T cell epitopes than DQ and DR.

[196] The epitopes 3-2018, 11 -2018, and 15-2018, which are strongly antigenic, bind to DPA1 *01:03/DPB1 *04:01 (Figure 15), indicating that a vaccine containing these epitopes could be used all over the world, not only because it binds to DPB1 *04:01 , but also because it binds to DPA1 *01 :03, which is the most frequent DPA allele in the world, and is prevalent in 71 .70% of the Brazilian population (Requena et al. 2020, Front. Immunol.1 1 , 2008, doi: 10.3389/fimmu.2020.02008). Coincidentally, DPA1 *01 :03 is present in exactly 7/10 patients analyzed (3 cured and 4 DTH+) (Figure 20). [197] Additionally, DPA1 *01:03 e is combined with DPB1 *04:01 , in five of them and with DPB1 *04:02 in 2 of them, matching the description that these two alleles are present in 60% of the human population.

[198] On the other hand, the DPB1 *04:02 allele, to which the epitopes 3-2018, 1 1 - 2018 and 2-2014 also bind (Figure 14), was recently described as prevalent in 64.98% of the population. Brazilian population (Requena et al., 2020). Therefore, a vaccine containing these epitopes could generate high efficacy. However, linking to only these highly prevalent alleles may not be interesting, as promiscuity is also important and optimizes vaccine efficacy.

[199] In contrast, with regard to DR, the DRB1 *03:01 allele, binds only to the 13-2018 epitope, and is present in 6.49% of the Brazilian population. On the other hand, the DRB1 *07:01 allele, which binds to the epitopes 3-2018, 9-2018, 1 -2014, 2-2014, 5-2018, 13-2018 and BL1, is present in 9.58% ; and the DRB1 *11:01 allele, which binds to 3-2018, 1 -2014, 5-2018, in 5.91% of the Brazilian population (Requena et al. 2020, Front. Immunol.1 1 , 2008, doi : 10.3389/fimmu.2020.02008) (Figure 15). The results revealed for the NH36 epitopes an interesting coverage profile of HLA class II molecules alleles.

Identification of binding sequences of class I HLA molecules within the epitopes for the NH36 class.

[200] In the ICS bioassay it was shown that NH36 epitopes, despite being predicted to bind MHC class II molecules, may also be binding to MHC class I molecules. In effect, they increased lymphocyte frequencies T CD8+ that express cytokines (Figure 9-12). To confirm this hypothesis, the epitope sequences for MHC Class II of NH36 were submitted to analysis by the “MHC Class I Binding predictions” program of the IEDB Database, using the NetMHCpan EL 4.1 program as a basis (Figure 16). The presence of possible linker sequences of HLA-A and HLA-B molecules was quantified. [201] The following were obtained: 1) the predicted number of linker sequences within each epitope, 2) the number of alleles associated with these sequences, and 3) the number of promiscuous linker sequences within each epitope (Figure 16). Additionally, Figure 16 shows the prediction of which of the 27 studied alleles (16 HLA-A* alleles and 11 HLA-B* alleles) these linker sequences might associate. The predominance of epitopes 11 -2018 and 13-2018 in these 3 parameters is notorious, followed by 15-17-2018, BL2 and 5-2018.

Analysis of promiscuity and population coverage

[202] Finally, analysis of the worldwide population coverage of epitopes for HLA class I and class II alleles was performed (Figure 17). This figure summarizes the population coverage for class I A and B loci alleles for the world population, given by each epitope studied. It was observed that 11 of the 15 epitopes studied have the capacity to bind to more than 50% of the class I alleles of the world population (Figure 17).

[203] The epitopes 11 -2018 and 13-2018, are the ones with the highest coverage for class I molecules, and together with the epitopes 2-2014, 5-2018, 7-2018 and BL1 , the ones with the highest coverage for molecules of class II (Figure 17). Indeed, the worldwide population coverage of epitope 11 -2018 is 93.9% and 92.7% for class I and class II, respectively (Figure 17). That of the 13-2018 epitope is 77.8% and 85.9% for class I and class II, respectively (Figure 17). The worldwide population coverage curves of the epitopes for Class I and Class II are shown in Figures 18 and 19, respectively.

[204] Below the curves, a smaller table shows coverage for the entire group of 15 epitopes. Individual epitopes have a wide margin of effectiveness, with coverage for Class I loci ranging from 7.8 to 93.9% (Figure 17). As a whole, the entire panel provides coverage for >98% of individuals (Figure 18), with an average individual recognizing approximately 11 different HLA epitope/molecule combinations. [205] This number is, however, an underestimate, since multiple Class I epitopes predicted for the same allele within the same CD4 epitope are not quantified. In fact, each allele is only quantified once.

[206] Epitope 1 1 , for example, has a Class I population coverage of 93.9% (Figure 17). This means that 93.9% of individuals are expected to have at least one Class I allele predicted to bind in a sequence included in epitope 11. However, prediction for binding of MHC molecules does not necessarily mean the ability to induce a T cell response. In addition to its impressive population coverage, it is also possible that other alleles may bind to other sequences contained in the 11 epitope.

[207] This result is very good, as the panel not only collectively provides broad coverage, ie that the epitopes all bind a high percentage of individuals, but also provides "deep" coverage (Figure 17). By depth it is meant that not only is the individual covered, but that he is covered by multiple different epitopes. This is important in cases where, for example, a dominant epitope for A*02:01 undergoes a mutation that causes the pathogen to now be able to evade the immune response. If there is deep coverage, the likelihood of evading detection and immune response is minimized, as multiple epitopes that bind to the A*02:01 allele, for example, are unlikely to mutate similarly at the same time.

[208] Figure 18 shows the population coverage curve for Class I. Each individual in a population will recognize a certain number of epitopes nested within an NH36 CD4 epitope. For example, someone who is A*02:01, A*03:01, B*07:02, and B*40:01 might recognize 4, 2, 1, and 5 epitopes restricted to those alleles, respectively; overall, they would recognize 12 different HLA/epitope combinations (4 for A*02:01, 2 for A*03:01, etc.). The bar chart shows the number of combinations "recognized" by different fractions of the population. Here, just over 5% of subjects (y-axis 1 ) will recognize 6 combinations (x-axis, bars), 7.5% will recognize 14, and about 2.5% will recognize 20, etc. [209] Cumulatively, adding from right to left, in 90% of individuals, it is expected that at least 3.19 epitopes will be recognized (PCgo, Figure 18). Likewise, about 75% of the population would recognize 6 or more, 50% about 12 or more, etc.

[210] Figure 17 also shows the population coverage for Class II (last gray column). It is important to point out, however, that the coverage for the prediction of class II molecules is underestimated because the DRB3/4/5 alleles have not been included. There are no reliable data on these alleles. Coverage is based on DRB1 , DQB1 and DPB1. Alternatively, additional coverage can be estimated based on the bond. That is, DRB4 is always associated with DRB1 *04, 07 and 09. So, for example, the peptides that bound DRB1 *04 can be added to the cover, assuming that individuals who have DRB1 *04 will express DRB4*. The same goes for DRB3 and DRB1 *03, 11 , 12, 13 and 14, and for DRB5 and DRB1*15 and 16.

[21 1] Figure 19 shows the proportion of individuals in the population that will recognize an epitope for a class II molecule. Bar 16 (x-axis) shows that 5.9% of subjects are predicted to recognize 16 different combinations of HLA class II peptides/molecules. Furthermore, 5.25% will recognize 10 combinations, 6% will recognize 13 combinations, etc. The sum of all columns will reach 100%, but it is possible that there is rounding, and that part of the population studies described in AFND do not necessarily correspond to all alleles in the population.

[212] The 57% cumulative coverage (y-axis 2) associated with column 16 means that 57% of the population will recognize 16 or more combinations of HLA class II peptides/molecules; 84% will recognize 10 or more, and 10% will recognize 28 or more, etc. The distribution, especially in the world population, is Gaussian. Smaller, less diverse populations may have small, less diverse coverage that may be skewed, but is usually normal. Additionally, it is expected that at least 7.67 of the epitopes will be recognized by 90% of the world's population (PC ₉₀ , Figure 19). [213] Coverage for NH36 epitopes is slightly higher for Class II (99.44, PC 7.67, Figure 19) than for Class I (98.55, PC 3.19, Figure 18). But the coverage is very good in both cases.

[214] Consequently, the results of the population coverage analysis support the selection of NH36 epitopes to produce potentially promising multi-epitope vaccine proteins for inclusion in an immunogenic composition against visceral Leishmaniasis that could be used universally.

Identification of epitopes capable of binding to HLA molecules correlated with susceptibility or resistance to VL

[215] This type of approach privileges epitopes that bind to HLA alleles that are highly associated with naturally resistant/protective, high-risk, or intermediate-risk phenotypes of contracting VL. leishgen et al., (2013) and SINGH et al., (2018) described that the DRB1 *15:01 and DRB1 *01 : 01 alleles are associated with the protective phenotype (Figure 15 and Table 5). We observed that 1 1 epitopes bind, in vitro, to DRB1 *01 : 01 and 10 to DRB1 *15: 01 alleles. 3-2018 and BL1 bind to both alleles. Furthermore, epitopes 1 -2014, 2-2014 and BL2 bind to DRB1 *15:01 , while epitopes 3-2014 and 9-2018 also bind to DRB1 *01:01 .

[216] On the other hand, six epitopes (1 -2014, 2-2014, 7-2018, 9-2018, 1 1 -2018 and 13-2018) also bind to the DQB1 *05:01 allele, which is associated with the intermediate risk of VL. Finally, eleven epitopes (9-2018, 1 1 -2018, 15-2018, 2-2014, 1 - 2018, 5-2018, 13-2018, 17-2018, 7-2018, BL1, BL2) bind , in vitro, to the DQB1 *02:01 allele, which is associated with a high risk of VL (LEISHGEN et al., 2013; SINGH et al., 2018) (Table 6 and Figure 15).

Table 6. NH36 epitopes that bind to alleles associated with risk and protection against VL (LEISHGEN et al., 2013)

[217] More recently, in 2018, the correlation of DRB1 alleles with risk or protection against VL was better detailed (Singh T et al. The Journal of Immunology, 2018, and Singh B et al Immunology 2018). Data are summarized in Table 7. In the first row of Table 7, the DRB1 alleles related to PROTECTION, INTERMEDIATE or HIGH RISK of VL were observed. In the second row, we observed the alleles, with 4-digit markings, that bound in vitro to the NH36 epitopes.

Table 7. NH36 epitopes that bind to DRB1 alleles associated with risk and protection against VL (Singh T et al., 2018)

[218] Results from the literature indicate that the NH36 epitopes studied are associated with both Class II HLA molecules associated with resistance or protection, and with high or intermediate risk of developing VL. According to Singh et al., 2018, some epitopes are only associated with protection, such as epitope 11 -2018 while others, such as 1 -2014, 2-2014 are associated with protection, intermediate or high risk of VL. On the other hand, according to LEISHGEN et al., 2013, the epitope 11 -2018 binds to alleles associated with protection, intermediate or high risk of VL.

[219] This analysis reinforces the importance of using the studied epitopes of NH36 for the formation of multi-epitope proteins for the production of an immunogenic composition (vaccine) that can protect not only individuals with phenotypic profile in HLA of natural resistance or protection, but also those with susceptibility profile or risk of contracting VL.

HLA typing of the patients studied

[220] DNA samples from 8 sick and cured patients and from 10 asymptomatic DTH+ individuals were obtained. The typing of 1 1 HLA loci alleles (HLA-A*, HLA-B*, HLA-C*, DRB1*, DBQ1*, DRB3*, DRB4*, DRB5*, DQA1*, DPA1*, DPB1*) was carried out using the massive parallel sequencing method, at the HLA-UERJ Laboratory, by the team of Professor Luiz Cristóvão de Moraes Sobrino Porto. The results obtained from sick patients who cured (cured) and asymptomatic DTH+ (DTH+) are summarized in Figure 20.

[221 ] The times in which each allele appears for each of the HLA loci was then quantified and we compared, by x ² assay with Yates correction (scipy.stats.chy.square contingency library in Phyton), whether the frequencies for a given allele were significantly different in cured patients and in asymptomatic DTH+ individuals.

[222] The HLA-C*04:01 allele was observed to be found in 6 of 8 cured individuals, which represent individuals who became ill, but in none of the DTH+. This difference was highly significant (p = 0.0043) indicating that the HLA-C*04:01 allele is correlated with susceptibility to human VL. This result is very relevant considering that the sample of cured and DTH+ individuals studied is small. This correlation had not been described in the literature before.

[223] On the other hand, studies by Leishgen, (2013) and Singh et al., (2018) revealed that the DRB1 gene concentrates the basis of susceptibility and natural resistance to VL. Indeed, the DRB1 *1501 , DRB1 *1502, DRB1 *01 :01 and DRB1 *16 alleles are associated with protection against VL, the DRB1 *03:01 , DRB1 *04, DRB1 *07:01 and DRB1 *1202 alleles has a neutral profile and the alleles DRB1 *10:01 , DRB1 *14:04, DRB1 *1 1 , DRB1 *13:01 , DRB1 *13:02 and DRB1 *09:01 with a high risk of contracting VL (Table 7 ).

[224] Although the sample of patients studied was relatively small, it was possible to observe that the alleles associated with protection DRB1 *0101 and DRB1 *1501 if considered together, are present in 4 of 10 DTH+ patients, however, they are not found in any of the 8 cured patients (x ² one tailed P value = 0.0023) (Figure 20).

[225] Thus, even the DRB1 *1301 allele linked to high risk of contracting VL is present in one patient and DRB1 *1302 in another cured patient, but neither appears in DTH+ individuals (Figure 20). In contrast, the DRB1 *10:01 and DRB1 *1 1 alleles, also associated with high risk, appear in 1 cured individual, each, but also in 2 and 3 DTH+ individuals, respectively.

[226] It is interesting to observe that 3 DTH+ patients (CY:5, CY:7, CY:9) present in their phenotype the DRB1 *01 :01 allele, related to protection against VL (Table 6) but also, the DQB1 allele *05:01 , pointed out by the work of Leishgen (2013), as related to intermediate risk. The CY:5 patient responded, as expected, to epitopes 1 to 1 1 , 15 and 17 of 2018, increasing the frequencies of IL-2 ⁺ TNF-α ⁺ IFN-γ ⁺ multifunctional CD4 T cells.

[227] The CY:7 patient responded most strongly to the 9-2018 epitope, and the CY:9 patient to the 1 1 -2018 epitope. In contrast, patients BY10 and BY:9, from the infected and cured group, had the DQB1 *02:01 phenotype associated with high risk. Patient BY:10 e responded to epitopes 2-2014, 17-2018, in addition to 3-2018 while patient BY:9 responded to epitopes 1 1 -2018, 9-2018 and 13-2018. Thus, the most relevant epitopes for the study sequence were selected.

Genetic modification of epitopes, multiepitope proteins and immunogenic composition

[228] Considering the strong antigenicity of NH36 epitopes, its ability to bind antibodies, induce cytokine secretion by CD4 and CD8 T lymphocytes, promote increased frequencies of central memory and effector lymphocytes, its promiscuity predicted by bioinformatics and proven in vitro, their broad population coverage on HLA class II and class I molecules, and their predicted strong binding to more specific alleles associated with protective, high-risk, and intermediate-risk responses against human VL, were therefore included the studied epitopes, in two formulations. [229] Although some of them have been highlighted as more potent inducers of immune memory, promoting higher frequencies of multifunctional T lymphocytes, it was decided to include all epitopes, favoring their promiscuity and wide population coverage that will allow them to protect more vulnerable populations. broad and diverse and that exercise universal protection against leishmaniasis.

[230] Even considering that the identified epitopes are indeed very immunogenic, the different types of arrangements that the epitopes can undergo within the multiepitope vaccine, can affect their potency. For example, they could be less efficient when generating junctional epitopes. The creation of junctional epitopes is a concern when designing linear polypeptides. A junctional epitope is defined as a neoepitope, which arises from the juxtaposition of two authentic epitopes.

[231] The new epitope is composed of the C-terminal sector of an epitope and the N-terminal sector of an epitope later in the sequence. The presence of a junctional epitope can create unwanted immunodominance effects, redirecting the immune response against irrelevant epitopes and, on some occasions, even suppressing the induction of a response against more desired epitopes (Livingston et al. J Immunol 2002;168:5499 -5506). To avoid the emergence of neoepitopes, a spacer sequence is placed between the predicted epitopes.

[232] The GPGPG sequence was designed by Livingston et al. (2002). It allowed restoration of the immunogenicity of HLA class II epitopes by disrupting the junctional epitopes that had formed. The sequence GPGPG was then considered as a universal spacer. On the other hand, Kolla et al. (2021) used AAY spacers to join cytotoxic T lymphocyte (CTL) epitopes. The AAY spacer is the cleavage site in the proteasome in mammalian cells. Kolla et al., (2021) also used the GPGPG sequence to join helper T lymphocytes and the KK spacer to join B lymphocyte epitopes, respectively, in a vaccine against Staphylococcus aureus. [233] Immunoinformatics studies demonstrated that the vaccine would be able to increase antibody responses, secreted levels of IL-2, IFN-γ, TGF-[3, and T and B cell responses, and induce immune responses primary, secondary and tertiary (Kolla et al., 2021 ). The KK spacer has also been used to enhance the antibody response in a nasal vaccine (Yano et al., 2003).

[234] KK sequences are the cut-off point for cathepsin, one of the important antigen processing enzymes within the HLA class II context. Additionally, Moradi et al., (2017) designed a DNA vaccine against murine tuberculosis, based on predicted epitopes for MHC class I, linked by AAY, and epitopes for HLA class II linked by GPGPG motifs, joined in tandem [162 ], Finally, in the patent application by McLeod R et al., 2017 (WO 2017/184456 Al) which deals with a multiepitope vaccine against toxoplasmosis, in addition to the epitope PADRE, the sequences of three alanines (AAA or KAAA ) and GPGPG as spacers.

[235] In the present invention, multiepitope vaccines are composed of genetically modified epitopes from the NH36 sequence. The originally studied epitopes were predicted for MHC Class II therefore they should be separated by GPGPG sequences. However, they contain within each one important MHC Class I sequences that should be separated by AAA-derived sequences. In order not to alter the structure of these important and potent epitopes, the strategy included the use of genetically modified epitopes from the NH36 sequence with the inclusion of AAA and GPGPG spacers, represented by SEQ ID NO:16 to SEQ ID NO:30 and SEQ ID NO:32 to SEQ ID NO:46, respectively.

[236] From these epitopes, it was possible to construct the genetically modified multiepitope proteins MultiAAA and MultiGPGPG. The formation of the MultiAAA protein occurred by joining the epitopes from SEQ ID NO:16 to SEQ ID NQ:30 and is represented in SEQ ID NO:31. The formation of the MultiGPGPG protein occurred by joining the epitopes from SEQ ID NO: 32 to SEQ ID NO: 46 and is represented by SEQ ID NO: 47. [237] The immunogenic composition (multiepitope vaccine) consists of at least one of the multiepitope proteins and saponin adjuvant in sterile 0.9% NaCl saline.

[238] In the case of the MultiAAA vaccine (SEQ ID No: 31 and Figure 21 ), the artificial sequences that compose it are modified by the addition of three Alanine residues at the C-Terminal ends, as described in Table 8. Only epitope 15 -17- 2018 which is modified by the addition of KRIGD, does not require three alanines as it is found at the end of the sequence, and will be followed by the 6 Histidine tail which facilitates the purification of the expressed peptide. As explained, the epitope 15-17-2018A (SEQ ID NO:28) replaces the epitopes 15-2018A (SEQ ID NO:26) and 17-2018A (SEQ ID NO:27) in the MultiAAA vaccine.

Table 8. Genetically modified and synthetic epitopes to compose the MultiAAA vaccine

[239] The Protparam-Expasy program revealed that the MultiAAA vaccine is composed of 310 amino acids and has a molecular weight of 32205.23 Daltons and a theoretical pl of 6.15. It has 29 negatively charged amino acid residues (Asp+Glu) and 23 positively charged residues (Arg+Lys). Additionally, the multiAAA vaccine has an estimated half-life of 20 h in mammalian reticulocytes in vitro, 30 minutes in yeast in vivo and over 10 hours in Escherichia coli in vivo. Its instability index is 24.38, which classifies it as a stable protein. Its aliphatic index is 107.13 and its average hydropathic index (GRAVY) is 0.422.

[240] In the case of the MultiGPGPG vaccine (SEQ No ID:47 and Figure 22), the artificial sequences that make up this multiepitope vaccine are modified by addition of GPGPG residues at the C-Terminal ends, as described in Table 9. Only the epitope 15-17-2018G (SEQ NO:44) which is modified by the addition of KRIGD, does not require the addition of GPGPG as it is found at the end of the sequence, and will be followed by the 6 Histidine tag that facilitates the purification of the expressed peptide. As explained, the 15-17-2018G (SEQ ID NO:44) epitope replaces the 15-2018G (SEQ ID NO:42) and 17-2018G (SEQ ID NO:43) epitopes in the MultiGPPGG vaccine.

Table 9. Genetically modified and synthetic epitopes to compose the multiGPGPG vaccine

[241] The Protparam-Expasy program revealed that the multiGPGPG vaccine is composed of 334 amino acids and has a molecular weight of 34031.06 Daltons and a theoretical pl of 6.15. It has 29 negatively charged amino acid residues (Asp+Glu) and 23 positively charged residues (Arg+Lys). It has an estimated average lifetime in mammalian reticulocytes in vitro of 20h, in yeast in vivo of 30 min and in Escherichia coli in vivo greater than 10h. Its instability index is 20.96, which classifies it as a stable protein. Its aliphatic index is 88.65 and its average hydropathic index (GRAVY) is 0.039.

[242] The MultiAAA and MultiGPGPG proteins were cloned into E. coli BL21 (DE3) and their expression induced in different concentrations of IPTG, as shown in Figure 25. When compared with the lysate of non-induced bacteria, the expression of the multiAAA protein is very clear, in the three concentrations of IPTG, showing an intense band corresponding to pm = 32.6kDa. On the other hand, the MultiGPGPG protein also showed to be a very intense band, in response to induction with the three concentrations of IPTG at molecular weight of 34.5 kDa.

[243] AAA (SEQ NO:31) and GPGPG (SEQ NO:47) multiepitope proteins are incubated with PBMC from pretreatment VL patients, asymptomatic DTH+ patients, and cured patients as well as PBMC from healthy controls. The cellular immune response generated in these patients will be analyzed by analyzing the labeling of intracellular cytokines by flow cytometry and the secretion of cytokines in the supernatants. The humoral antibody response in serum will also be analyzed. These data will allow evaluating the immunogenicity and antigenicity of the MultiAAA and MultiGPGPG vaccines against human visceral leishmaniasis.

[244] Analysis of the cellular immune response of PBMC from asymptomatic patients to MultiAAA multiepitope protein

[245] PBMC from asymptomatic patients were incubated, in the absence of stimulation or in the presence of purified recombinant MultiAAA protein, at concentrations of 12.5 pg/ml, 25 pg/ml, or 50 pg/ml, or with SLA at a concentration of 10 pg/ml (Figure 26). It was possible to observe a dose-response frequency of IL-2 secreting CD4+ T lymphocytes whether mono-secretors or total frequencies. On the other hand, the frequencies of mono-secretory CD4+ T lymphocytes, or total frequencies of IFN-y secretors, are maximal in response to a concentration of 25 pg/ml of the protein (Figure 26). It is notorious the greatly increased frequencies, close to 12%, in response to a concentration of 50 pg/ml. These results indicate that the multiepitope vaccine design potentiated about 60 times the immunogenicity observed for the NH36 domains and for the NH36 epitopes, separately. Likewise, it is observed that the response induced by the AAA multiepitope protein is much higher than that of the SLA positive control.

[246] Adjuvants for immunogenic composition and Fasel and 2 assays

[247] The term adjuvant comes from the Latin adjuvare, which means to help, consisting of compounds added to immunogenic antigens that potentiate the immune response against them. These compounds have been used to improve the efficacy of vaccines since the 1920s. The search for new adjuvants has been increasing in recent years, as they increase the efficacy of purified, recombinant or synthetic antigens.

[248] For multiepitope protein immunogenic compositions, saponin QS21 alone or in combination with Monophosphoril is used as an adjuvant. Salmonella minessota Lipid A (MPL), or the ISCOMs particulate adjuvants of the ISCOMATRIX type, which are also based on saponin QS21.

[249] Saponin QS21 is the most commonly used saponin and is a product of Quillaja saponaria Molina. They are complex molecules of high molecular weight, which are in the form of natural conjugates of triterpenes, steroids or steroidal glycoalkaloids with one or more sugar chains. The term saponin refers to its classic detergent property, which in turn is related to the antipathetic nature of the molecule (a glycidic moiety that is hydrophilic and an aglycon or sapogenin that is hydrophobic). They are substances widely used in veterinary and human vaccines.

[250] Leishmune® vaccine adjuvant contains acylated and deacylated saponins QS21 as the active component. They stand out for their ability to induce a potent Th1 response, increase levels of lgG2a, IFN-y and IL2. One of the proposed mechanisms of action is that saponins intersperse in cell membranes through their hydrophobic fraction, causing pore formation, which facilitates antigen presentation via the MHC-1 pathway and thus promotes increased CTL response. It has also been described that the presence of the aldehyde group on C-4 allows saponin to mimic the B7 ligand, increasing the stimulation of APCs on CD4 helper lymphocytes (TH1 ).

[251] Saponins from Quillaja saponaria, which are conjugates of two sugar moieties associated with C-3 and C-28 of a triterpene nucleus, have two peculiar characteristics that make them very potent: the presence of an aldehyde group (CHO) in the C-4 and the presence of normoterpenes in the C-28 of the triterpene. On the surface of antigen-presenting cells, the aldehyde group of the costimulatory molecule B7 interacts with amines on the surface of lymphocytes to form Schiff bases. This interaction triggers the transmission of a signal that stimulates the TH1 response.

[252] Thus, it has been proposed that the aldehydes of Quillaja saponaria saponins act as analogues of the family of costimulatory molecules, B7 ligands, on the surface of antigen-presenting cells, also independently forming Schiff bases. On the other hand, normonoterpene linked to C-28 is responsible for the toxicity and cytotoxic response (CD8) induced by treatment with saponins (Lacaille-Dubois, 2019). Saponin QS21 also acts through the activation of the inflammasome and the consequent release of cytokines dependent on caspase-1, such as IL-1 β and IL-18 which are important for the Th1 response. There are currently around 50 synthetic analogues of QS21. It has been used in human vaccines at doses of 50pg or 100pg.

[253] Additionally, QS21 has a synergistic effect when administered as MPL-A in a liposome (AS01 ). Liposomes are synthetic nanospheres consisting of phospholipid bilayers, which encapsulate antigens and act as a delivery system. MPL is a detoxified derivative of LPS from Salmonella minnesota, it induces activation of innate immunity through TLR-4 and can be classified as a PAMP. Like LPS, it acts on the Toll-4 receptor of professional antigen-presenting cells, which promote the release of IL2 and IFN-Y cytokines that induce a TH1 response. Several synthetic derivatives were generated according to the study of the function-structure of MPL, including RC-529 (Singh & O'Hagan, 2003).

[254] Several trials of immunizations including Leishmania, malaria, human papillomavirus (HPV), Hepatitis B virus (HBV), tuberculosis, and HIV with different formulations of MPL have been shown to be safe and effective with this adjuvant (Reed et al., 2009). This series of adjuvants was developed by Glaxo Smith Kline and is used in prophylactic vaccines against malaria, Herpes zoster, tuberculosis, AIDS and in therapeutic vaccines against cancer and Alzheimer's disease. Indeed, QS21 within AS adjuvant forms part of the Herpes Zoster (HZ/su) vaccine (Shingrix™) which was licensed in 2017 by the FDA and received marketing authorization in 2018, and the RTS,S/AS01 Mosquirix™ vaccine against Malaria which was approved by EMA in 2015 for implementation in Sub-Sahara. In children, Mosquirix is used containing 62.5 pg of antigen with 50 pg of MPL and 50 pg of Q21 in an oil-water emulsion finally diluted in 250 pl of saline.

[255] Finally, ISCOMs are 40 µm particles composed of cholesterol, phospholipids, and saponins from Quillaja saponaria Molina. For vaccines the ISCOMs also contain the desired antigen. This preparation reduces the toxicity of these saponins, decreasing their hemolytic action. In ISCOM, saponin is linked to cholesterol and does not interact freely with cell membranes. ISCOMs are established to induce Th1 and Th2 and pro-inflammatory cytokine responses and have been studied in several animal models and an equine influenza vaccine using ISCOMs is licensed in Switzerland. ISCOMs can be used through the oral, respiratory and vaginal routes and are administered as an adjunct to Merck Sharp and Dohme vaccines.

[256] In conclusion, in human vaccines, saponin QS21 alone is used at doses of 50 pg, and ISCOMs of the ISCOMATRIX type at doses of 100 pg to 120 pg (Bigaeva et al., 2016) with the adjuvant ASO1 containing 50 pg of QS21 and 50 pg of MPL-A (Regules et al., 2016). Both adjuvants were formulated with variable doses of antigens. It is worth remembering that in the Leishmune® vaccine against canine leishmaniasis, the adjuvant contains 156 pg of QS21 per dose (Oliveira-Freitas et al., 2006).

[257] Therefore, the optimal immunogenic composition (vaccine) of multiepitope proteins, object of this invention, is characterized by containing at least one multiepitope protein (multi AAA or multi GPGPG) in concentrations of 10 to 200 pg formulated with 50 pg of QS21 , or 100pg to 120pg of ISCOMATRIX, or with ASO1 adjuvant composed of 50pg of QS21 and 50pg of MPL-A. Each formulation will be hydrated in 0.2 to 0.5 ml of sterile 0.9% NaCI saline solution.

[258] In Phase 1 and 2 trials, formulations containing MultiAAA and MultiGPGPG multiepitope proteins will be prepared 1 hour before use and kept at 4°C until application. The absence of LPS from the preparations will be confirmed using the LAL QCL-1000 kit (Lonza). For safety and tolerability tests, the vaccines will be applied in three subcutaneous doses at intervals of 28 days (day 0, day 28 and day 56). The safety of the vaccines will be evaluated 1 h, 2 days and 7 days after each injection.

[259] Subjects will be monitored for adverse effects for up to one year after administration of vaccines. Adverse effects include injection site pain, swelling, redness and induration, fever, myalgia, malaise, fatigue, headache, diarrhea, anorexia, skin rash and allergic reactions. A grade 1 adverse effect score will be established: mild and not limiting activity, 2: affecting function but not activities of daily living, 3: severe, affecting activities of daily living, 4: severe, threatening life or cause disability. Vital signs will be assessed at each return visit and 30 min after each injection.

[260] For the toxicity test, blood samples will be taken to obtain sera, 1 h before and 6 h, 12 h and 24 h after the injection. In these sera, the values of alkaline phosphatase, gamma glutamyl transferase, sodium, Potassium, lipase, urea amylase, creatinine, prothrombin time, IMR, albumin/globulinin, C-reactive protein and electrocardiograms. Additionally, sera will be collected before injections and 30 days after to measure anti-nuclear antibodies and rheumatoid factor, as an indication of possible onset of autoimmunity.

[261] Finally, for the immunogenicity test; Blood samples and sera will be collected on days 0, before vaccine injection, and on days 30, 60 and 150 after the first vaccine dose. Sera will be evaluated by Elisa assay against multiAAA and multiGPGPG proteins and Leishmania (L.) infantum lysate (2 pg/well), using anti-IgG, IgG1, IgG2, IgG3 and IgG4 antibodies for development. PBMC will be purified in Fycoll gradient on day 0, before vaccine injection and on days 30, 60 and 150 after the first dose.

[262] PBMC will be plated with multiepitope vaccines and promastigotes lysate and incubated for 72 h at 30 °C and 5% CO2. The secretion of cytokines in the supernatants will be evaluated with the Human Th17 kit, 9 plex (IFN-g, TNF-αlpha, IL-1B, IL-2, IL-4, IL-6, IL-10, IL-12p70, IL-17A) - HTH17MAG-14K-09 and analyzed using the Milliplex Analyzer 5.1 software (Merck Millipore, Billerica, USA) and the Multispecies TGF[3 - Single Plex (TGF[31) kit.

[263] In addition, intracellular labeling of cytokines will be performed, cultivating the PBMCs in vitro with 10 pg / ml of promastigotes lysate, or multiAAA or multiGPGPG proteins, for 6 h, followed by the addition of Brefeldin A and an additional incubation for 12 H. Cells will be stained for surface markers CD3, CD4, and CD8, and later for the intracellular expression of cytokines IL-2, TNF-α and IFN-y. Events will be acquired on a BD FACSCanto II™ flow cytometer and data analyzed with FlowJo software (boolean gain).

[264] At 150 days after the first injection, after collecting blood and serum, an IDR with Leishmania lysate will be applied to vaccinated individuals and those treated with saline placebo, and the size of the induration will be evaluated up to 72 hours after injection. Kruskall-Wallis and Mann-Whitney tests will be used for comparison and Spearman's two-tailed correlation test will be used for correlation analysis using Graph Pad Prism 9 software.

Brief description of the Figures

[265] Figure 1. Analysis of serum humoral response to NH36 epitopes. Anti-IgG, IgG1, IgG2, IgG3 and IgG4 antibody levels against epitopes 1 -2014 (SEQ ID NO:1), 2-2014 (SEQ ID NO:2), 3-2014 (SEQ ID NO:3), 1-2018 (SEQ ID NO:4), 3-2018 (SEQ ID NO:5 ), 5-2018 (SEQ ID N0:6), 7-2018 (SEQ ID N0:7), 9-2018 (SEQ ID N0:8), 11-2018 (SEQ ID N0:9), 13-2018 ( SEQ ID NQ:10), 15-2018 (SEQ ID N0:11) and 17-2018 (SEQ ID NO:12) were evaluated by an ELISA assay using serum from asymptomatic VL-cured individuals (n=10). DTH+ (n=10) and healthy controls from a non-endemic area (n=7). Asterisks represent significant differences between the absorbances of patients and their respective healthy controls (ANOVA) with the Sidak test for multiple comparisons.

[266] Figure 2. Analysis of serum humoral response to NH36 epitopes. “Heatmap” representing the mean absorbances of each type and subtype of antibody in response to the epitopes: 1 -2014 (SEQ ID NO:1 ), 2-2014 (SEQ ID NO:2), 3-2014 (SEQ ID NO: 3), 1-2018 (SEQ ID NO:4), 3-2018 (SEQ ID NO:5), 5-2018 (SEQ ID NO:6), 7-2018 (SEQ ID NO:7), 9-2018 (SEQ ID NO:8), 11-2018 (SEQ ID NO:9), 13-2018 (SEQ ID NO:10), 15-2018 (SEQ ID NO:11) and 17-2018 (SEQ ID NO:12 ). The anti-lgG, IgG 1 , IgG2, IgG3 and IgG4 antibody response was evaluated in sera from individuals cured of VL (n=10), asymptomatic DTH+ (n=10) and healthy controls from a non-endemic area (n= 7).

[267] Figure 3. Gate strategy for analysis of multifunctional CD4+ and TCD8+ T lymphocytes. Initially, the population of viable cells can be observed (a), from this population the population of total lymphocytes was selected using the parameters FSC x SSC (b), within this population we selected only CD3+ T lymphocytes (c), and then the TCD3+ CD4+ and TCD3+TCD8+ lymphocytes (d), from these populations of CD4+ and CD8+ lymphocytes, those that were double positive for CCR7+ CD45RA+ (naive T cells) were excluded, and only effector and memory lymphocytes were selected ( e), and within this population we evaluated the production of pro-inflammatory cytokines IL-2, TNF-α and IFN-y (f) and their possible combinations through Boolean Gate analysis.

[268] Figure 4. Percentage of CD3+ lymphocytes (a), CD3 ⁺ CD4 ⁺ (b) and CD3 ⁺ CD8 ⁺ (c). These frequencies were evaluated in PBMC from healthy controls (n=10), cured VL patients (D180) (n=10) and DTH+ individuals (n=10) when challenged with the NH36 epitopes 1 -2014 (SEQ ID NO :1 ), 2-2014 (SEQ ID NO:2), 3-2014 (SEQ ID NO:3), 1-2018 (SEQ ID NO:4), 3-2018 (SEQ ID NO:5), 5- 2018 (SEQ ID NO:6), 7-2018 (SEQ ID NO:7), 9-2018 (SEQ ID NO:8), 11 -2018 (SEQ ID NO:9), 13- 2018 (SEQ ID NQ:10), 15-2018 (SEQ ID NO:11) and 17-2018 (SEQ ID N0:12) (25pg/ml), and Leishmania Soluble Antigen (SLA) 10|_ig/ml .

[269] Figure 5. Frequencies of CD4+ T lymphocytes secreting a single cytokine (IL-2, TNF-α, or IFN-y). These frequencies were evaluated in PBMC from healthy individuals (n=6), cured VL patients (D180) (n=10) and DTH+ individuals (n=10) when stimulated with the synthetic epitopes of NH36 1 -2014 (SEQ ID NO: 1 ), 2-2014 (SEQ ID NO:2), 3-2014 (SEQ ID NO:3), 1-2018 (SEQ ID NO:4), 3-2018 (SEQ ID NO:5), 5 -2018 (SEQ ID NO:6), 7-2018 (SEQ ID NO:7), 9-2018 (SEQ ID NO:8), 11 - 2018 (SEQ ID NO:9), 13-2018 (SEQ ID NO:9) :10), 15-2018 (SEQ ID NO:11) and 17-2018 (SEQ ID NO:12) (25pg/ml) and SLA (10pg/ml). Results are represented as means + SE. For statistical analysis, the 95% confidence interval test was used.

[270] Figure 6. Frequencies of CD4+ T lymphocytes secreting two cytokines (IL-2 and TNF-α; TNF-α and IFN-y or IL-2 and IFN-y). These frequencies were evaluated in PBMC from healthy individuals (n=6), cured VL patients (D180) (n=10) and DTH+ individuals (n=10) when stimulated with the synthetic epitopes of NH36 1 -2014 (SEQ ID NO:1 ), 2-2014 (SEQ ID NO:2), 3-2014 (SEQ ID NO:3), 1-2018 (SEQ ID NO:4), 3-2018 (SEQ ID NO:5), 5 -2018 (SEQ ID NO:6), 7-2018 (SEQ ID NO:7), 9-2018 (SEQ ID NO:8), 11 -2018 (SEQ ID NO:9), 13-2018 (SEQ ID NO: :10), 15-2018 (SEQ ID NO:11) and 17-2018 (SEQ ID NO:12) (25pg/ml) and SLA (10pg/ml). Results are represented as means + SE. For statistical analysis, the 95% confidence interval test was used.

[271] Figure 7. Frequencies of CD4+ T lymphocytes secreting three cytokines (IL-2, TNF-α and IFN-y). These frequencies were evaluated in PBMC from healthy individuals (n=6), cured VL patients (D180) (n=10) and DTH+ individuals (n=10) when stimulated with synthetic epitopes of NH36: 1 -2014 (SEQ ID NO:1 ), 2-2014 (SEQ ID NO:2), 3-2014 (SEQ ID NO:3), 1-2018 (SEQ ID NO:4), 3-2018 (SEQ ID NO:5), 5 -2018 (SEQ ID NO:6), 7-2018 (SEQ ID NO:7), 9-2018 (SEQ ID NO:8), 11 - 2018 (SEQ ID NO:9), 13-2018 (SEQ ID NO:9) :10), 15-2018 (SEQ ID NO:11) and 17-2018 (SEQ ID NO:12) (25pg/ml) and SLA (10pg/ml). Results are represented as means + SE. For statistical analysis, the 95% confidence interval test was used.

[272] Figure 8. Analysis of the multiparametric ICS response of TCD3+CD4+ cytokine-secreting lymphocytes after incubation with NH36 epitopes. “Heatmap” representing the mean frequencies of CD4+ T cells secreting one, two or three cytokines from individuals cured of VL (n=10), asymptomatic DTH+ (n=10) and healthy controls from a non-endemic area (n=7 ).

[273] Figure 9. Frequencies of CD8+ T lymphocytes secreting a single cytokine (IL-2, TNF-α, or IFN-y). These frequencies were evaluated in PBMC from healthy individuals (n=6), cured VL patients (D180) (n=10) and DTH+ individuals (n=10) when stimulated with synthetic epitopes of NH36: 1 -2014 (SEQ ID NO:1 ), 2-2014 (SEQ ID NO:2), 3-2014 (SEQ ID NO:3), 1-2018 (SEQ ID NO:4), 3-2018 (SEQ ID NO:5), 5 -2018 (SEQ ID NO:6), 7-2018 (SEQ ID NO:7), 9-2018 (SEQ ID NO:8), 11 - 2018 (SEQ ID NO:9), 13-2018 (SEQ ID NO:9) :10), 15-2018 (SEQ ID NO:11) and 17-2018 (SEQ ID NO:12) (25pg/ml) and SLA (10pg/ml). Results are represented as means + SE. For statistical analysis, the 95% confidence interval test was used.

[274] Figure 10. Frequencies of CD8+ T lymphocytes secreting two cytokines (IL-2 and TNF-α, TNF-α and IFN-y, and IL-2 and IFN-y). These frequencies were evaluated in PBMC of healthy individuals (n=6), cured VL patients (D180) (n=10) and DTH+ individuals (n=10) when stimulated with synthetic epitopes of NH36: 1 -2014 (SEQ ID NO:1 ), 2-2014 (SEQ ID NO:2), 3-2014 (SEQ ID NO:3), 1-2018 (SEQ ID NO:4), 3-2018 (SEQ ID NO:5), 5 -2018 (SEQ ID NO:6), 7-2018 (SEQ ID NO:7), 9-2018 (SEQ ID NO:8), 11 -2018 (SEQ ID NO:9), 13-2018 (SEQ ID NO:8) :10), 15-2018 (SEQ ID NO:11) and 17-2018 (SEQ ID NO:12) (25pg/ml) and SLA (10pg/ml). Results are represented as means + SE. For statistical analysis, the 95% confidence interval test was used.

[275] Figure 11. Frequencies of CD8+ T lymphocytes secreting IL-2, TNF-α and IFN-y. These frequencies were evaluated in PBMC from healthy subjects (n=6), cured VL patients (D180) (n=10) and DTH+ subjects (n=10) incubated in the presence of synthetic epitopes of NH36: 1 -2014 (SEQ ID NO:1 ), 2-2014 (SEQ ID NO:2), 3-2014 (SEQ ID NO:3), 1-2018 (SEQ ID NO:4), 3-2018 (SEQ ID NO:5), 5-2018 (SEQ ID NO:6), 7-2018 (SEQ ID NO:7), 9-2018 (SEQ ID NO:8), 11 -2018 (SEQ ID NO:9), 13-2018 (SEQ ID NQ:10), 15-2018 (SEQ ID NO:11) and 17-2018 (SEQ ID NO:12) (25pg/ml) and SLA (10pg/ml). Results are represented as means + SE. For statistical analysis, the 95% confidence interval test was used.

[276] Figure 12. Multiparametric ICS response analysis of TCD3+CD8+ cytokine-secreting lymphocytes after incubation with NH36 epitopes. “Heatmap” representing the mean frequencies of CD4+ T cells secreting one, two or three cytokines from individuals cured of VL (n=10), asymptomatic DTH+ (n=10) and healthy controls from a non-endemic area (n=7 ).

[277] Figure 13. Levels of cytokines produced by PBMC's of DTH+ individuals stimulated in vitro with the epitopes of NH36. PBMCs from asymptomatic individuals (n=6) were incubated with 25pg/ml of the NH36 epitopes: 1 -2014 (SEQ ID NO:1 ), 2-2014 (SEQ ID NO:2), 3-2014 (SEQ ID NO: :3), 1-2018 (SEQ ID NO:4), 3-2018 (SEQ ID NO:5), 5-2018 (SEQ ID NO:6), 7-2018 (SEQ ID NO:7), 9- 2018 (SEQ ID NO:8), 11-2018 (SEQ ID NO:9), 13-2018 (SEQ ID NO:10), 15-2018 (SEQ ID NO:11) and 17-2018 (SEQ ID NO: 12), with soluble Leishmania antigen (SLA) or without stimulus (control) for a period of 72 hours, and cytokine secretion was measured in the supernatant using the Luminex assay. For statistical analysis, the 95% CI test was used.

[278] Figure 14. Prediction of HLA class II molecules that bind to the 27 most common alleles of the DRB*1, DRB*3, DRB*4 and DRB*5, DQA*1/DQB*1 and DPA*1 genes / human DPB*1. The NH36 epitopes predicted for human HLA class II molecules were analyzed by the IEDB recommended 2.22 program with percentile rank < 10% (in yellow) and < 20 % (in orange). The results show the lowest percentile rank values with which the linker sequences within the epitopes bind the HLA molecules.

[279] Figure 15. In vitro affinity test of epitopes for HLA-class II of NH36 by the molecules of the 27 most common alleles of the genes DRB*1, DRB*3, DRB*4 and DRB*5, DQA*1/DQB *1 and human DPA*1/DPB*1. This test analyzes the ability of epitopes to inhibit the binding of a radioactive peptide probe to purified MHC molecules. IC50 concentrations < to 1000 ηM, represented in bold, indicate the highest affinities of an epitope for each HLA receptor. Negative signs indicate IC50 values > 30000 ηM. The green color in the DRB1 *01 :01 and DRB*1 15:01 alleles shows their association with the PROTECTOR phenotype; the pink color on the DQB1 *05:01 allele shows its association with the INTERMEDIATE RISK phenotype and the red color on the DQB1 *02:01 allele shows its association with the ELEVATED RISK for VL.

[280] Figure 16. Identification of the linker sequences of HLA class I molecules within the epitopes for class II of NH36. The epitope sequences for MHC Class II from NH36 were analyzed by the “MHC Class I Binding predictions” program from the IEDB Database, NetMHCpan EL 4.1. We obtained: 1) the predicted number of linker sequences in each epitope, 2) the number of alleles associated with these sequences, and 3) the number of promiscuous linker sequences within each epitope.

[281] Figure 17. Worldwide epitope population coverage for HLA class I and II molecules. Predictions were performed using the IEDB Database http://tools.iedb.org/population tool, considering the allele frequencies in the world population based on the “Allele Frequencies database” (http://www.allelefreguencies.net/). Columns on the right show the values of world population coverage for each epitope.

[282] Figure 18. Worldwide population coverage for HLA class I molecules. Predictions were performed using the IEDB Database Population coverage Class I separate tool (http://tools.iedb.org/population,) considering allele frequencies in the world population based on the “Allele Frequencies database”. The bar graph shows the number of sequence combinations and HLA molecules "recognized" by different fractions of the population. In 90% of individuals, it is expected that at least 3.19 epitopes will be recognized (PC90).

[283] Figure 19. Worldwide population coverage for HLA class II molecules. Predictions were performed using the IEDB Database tool Predictions were performed using the IEDB Database tool Population coverage, Class II separate (http://tools.iedb.org/population). The bars show what proportion of individuals in the population will recognize an epitope for a class II molecule. The sum of all columns will add up to 100%. Additionally, it is expected that at least 7.67 of the epitopes will be recognized by 90% of the world's population (PC90).

[284] Figure 20. DNA typing of the patients studied for the genes HLA-A,B,C, HLA DQA and DQB, HLA-DPA and DPB and HLA-DR1, DR3, DR4 and DR5. DNA samples from DTH+ or cured VL patients were extracted from blood and/or bone marrow. The alleles are described with 4 digits and were identified by the massive parallel sequencing method. Alleles in green are associated with protection and natural resistance against VL. Alleles in red are associated with a high risk of contracting VL. [285] Figure 21. Molecular model of the multi-AAA multiepitope vaccine. The vaccine is composed of the epitopes genetically modified through the addition of AAA spacers (SEQ NO ID:31). The molecular model was obtained with the Swiss modeler and Pymol programs, using the pdb 1 EZR (Crystal Structure of Nucleoside hydrolase from Leishmania major. doi: 10.2210/pdb1 EZR/pdb) model as a base. The MultiAAA protein sequence with the AAA spacers in red. The table assigns colors to each artificial sequence or epitope modified by the addition of three Alanines.

[286] Figure 22. Molecular model of multiepitope multiGPGPG vaccine. The vaccine is composed of the genetically modified epitopes through the addition of GPGPG spacers (SEQ NO ID:47). The molecular model was obtained with the Swiss modeler and Pymol programs, using the pdb 1 EZR (Crystal Structure of Nucleoside hydrolase from Leishmania major. doi: 10.2210/pdb1 EZR/pdb) model as a base. The MultiGPGPG protein sequence with the GPGPG spacers in red. The table assigns colors to each artificial Sequence or epitope modified by the addition of the sequence Glycine-Proline-Glycine-Proline-Glycine.

[287] Figure 23. Schematic representation of MultiAAA multiepitope vaccine. The vaccine is composed of 12 sequences genetically modified by the addition of the AAA spacer and a last epitope linked to a tail of 6 Histidines. Each sequence represents an epitope for helper CD4+ T cells, which also contains epitopes for CD8+ T cells.

[288] Figure 24. Schematic representation of MultiGPGPG multiepitope vaccine. The vaccine is composed of 12 sequences genetically modified by the addition of the spacer GPGPG and a last epitope linked to a tail of 6 Histidines. Each sequence represents an epitope for helper CD4+ T cells.

[289] Figure 25. Expression of cloned MultiAAA and MultiGPGPG proteins in E coli BI21 (DE3). SDSPAGE results represent the profile of bacterial lysates without or with induction of expression with 0.5 mM, 0.75 mM or 1 mM IPTG.

[290] Figure 26. Frequencies of IL-2 mono-secretor CD4+ T lymphocytes, or IFN-y, and IL-2 and IFN-y dual-secretors. These frequencies were evaluated in PBMC of an asymptomatic individual before the stimulus with the recombinant multiepitope protein MultiAAA from NH36, composed, from the N end to the C-terminus, by the epitope sequence 1 -2018A(SEQ ID NO:19), 3-2018A (SEQ ID NO:20), BL1A (SEQ ID NO:29), 1-2014A (SEQ ID NO:16), 2-2014 ( SEQ ID NO:17), 5-2018A (SEQ ID NO:21), BL2A (SEQ ID NO:30), 3-2014A (SEQ ID NO:18), 7-2018A (SEQ ID NO:22), 9 -2018A (SEQ ID NO:23), 11 -2018A (SEQ ID NO:24), 13-2018A (SEQ ID NO:25) and 15-17-2018A (SEQ ID NO:28) at concentrations of 12.5 pg/ml, 25 pg/ml and 50 pg/ml). As a control, PBMC were incubated in the absence of antigenic stimulus or with SLA (10pg/ml).

Claims

1. Genetically engineered NH36 epitopes characterized by the amino acid sequences of SEQ ID NO:16 or SEQ ID NO:17 or SEQ ID NO:18 or SEQ ID NO:19 or SEQ ID NO:20 or SEQ ID NO:21 or SEQ ID NO:22 or SEQ ID NO:23 or SEQ ID NO:24 or SEQ ID NO:25 or SEQ ID NO:26 or SEQ ID NO:27 or SEQ ID NO:28 or SEQ ID NO:29 or SEQ ID NQ :30 or SEQ ID NO:32 or SEQ ID NO:33 or SEQ ID NO:34 or SEQ ID NO:35 or SEQ ID NO:36 or SEQ ID NO:37 or SEQ ID NO:38 or SEQ ID NO:39 or SEQ ID NO:40 or SEQ ID NO:41 or SEQ ID NO:42 or SEQ ID NO:43 or SEQ ID NO:44 or SEQ ID NO:45 or SEQ ID NO:46.

2. Genetically modified epitopes of NH36, as described in claim 1, characterized by the genetic modification occurring by the addition of spacers in the terminal portion of the epitope being three residues of Alanine (AAA) or residues of Glycine-Proline-Glycine-Proline-Glycine (GPGPG ).

3. Genetically modified multiepitope protein characterized by the amino acid sequence of SEQ ID NO:31.

4. Multiepitope protein, as described in claim 3, characterized in that it is formed by joining the epitopes from SEQ ID NO:16 to SEQ ID NQ:30 linked by AAA spacers containing one or two tails of 6 terminal histidines.

5. Genetically modified multiepitope protein characterized by the amino acid sequence of SEQ ID NO:47.

6. Multiepitope protein, as described in claim 5, characterized in that it is formed by joining the epitopes from SEQ ID NO:32 to SEQ ID NO:46 linked by GPGPG spacers containing one or two tails of 6 terminal histidines.

7. Multiepitope proteins, as described in claims 3 to 6, characterized by the junction of epitopes occurring regardless of sequential order, with the possibility of repetition of one or more epitopes.

8. Immunogenic composition characterized by containing at least one multiepitope protein, as described in claims 3 to 6, in concentrations of 10 to 200 pg, adjuvants in concentrations of 50 to 120 pg, hydrated in 0.2 to 0.5 ml of solution sterile 0.9% NaCl saline.

9. Use of the composition as described in claim 8, characterized in that it is to prepare a recombinant protein or DNA or synthetic vaccine for the prevention, treatment or diagnosis of human and animal leishmaniasis.